yennanliu avatar

us-stock-analysis

Comprehensive US stock evaluation: fundamental, technical, economic, multi-method valuation, financial report analysis, forensic accounting, credit check, head-to-head compare, ETF analysis, earnings preview, thesis tracking, risk stress test, tax lens, fact-check, options, dividends, insider trading, institutional ownership, bear-case red-teaming, position laddering, trade post-mortem, and more

version
1.15.0
GitHub stars
286

Install

/plugin install us-stock-analysis@invest-skill

Run it inside Claude Code after adding the marketplace with /plugin marketplace add yennanliu/InvestSkill.

Contents

Discovered from source

Skills 39

10-K Report Digest

⚠️ Data Verification — Do This Before Any Analysis

Before running the digest, confirm the filing details:

  1. Identify the filing — confirm ticker, fiscal year-end, and exact 10-K filing date from SEC EDGAR.
  2. State your source — note where the document came from (e.g., "SEC EDGAR, FY2024 10-K filed 2025-02-05") at the top of the output.
  3. Flag incomplete data — if only partial sections are available, list what is missing before proceeding.

⚠️ Source unavailable. If the full 10-K cannot be retrieved, display this warning and proceed only with the sections provided.


📋 Data & Sources Header — Open Every Output With It

The first thing in the output is this provenance block, filled in — never left as placeholders. It is the standard documented on the Data & Accuracy page and the first thing result-validator looks for:

Data & Sources
  As of:      <date the figures represent, e.g. 2026-06-30>
  Source:     <primary docs — SEC EDGAR 10-K/10-Q, company IR, FRED, exchange data …>
  Retrieval:  <pasted by user | web/tool retrieval | model memory>
  Confidence: <HIGH | MEDIUM | LOW>
  • Retrieval: model memory must be paired with Confidence: LOW — memory is a placeholder until confirmed against a primary source.
  • Mixed sources: list each with its own as-of date rather than blending them.
  • Sources disagree on the same figure (a closing price, a deal term): prefer the primary source (SEC filing, company release, exchange data) over an aggregator, then the most recent dated one. Show both values with their sources and dates, say which you used and why, and never average them.
  • Data the user pasted is reported as pasted by user; do not upgrade its confidence beyond what the user's own source supports.

Produce a clean, structured markdown document that distills a company's 10-K annual report into an abstract, per-section summaries, a digest of key metrics, and full source references. Designed for quick comprehension — not trading signals. Output language is selectable: English or Traditional Chinese (繁體中文).

🔎 Retrieving the Filing — the Keyless EDGAR Path

This skill fetches nothing by itself. When the user gives only a ticker, obtain the 10-K in one of three ways and record which one in the Data & Sources header:

A. The host can fetch URLs → Retrieval: web/tool retrieval. Every step is free and needs no API key:

  1. Ticker → CIK — https://www.sec.gov/files/company_tickers.json: find the entry whose ticker matches (write BRK-B for BRK.B) and zero-pad its cik_str to 10 digits.
  2. Filings index — https://data.sec.gov/submissions/CIK##########.json: inside filings.recent the arrays form, filingDate, reportDate, accessionNumber and primaryDocument are aligned by index. Take the newest row whose form is 10-K (foreign private issuers file 20-F instead; a 10-K/A is an amendment). The top-level fiscalYearEnd is MMDD.
  3. The document — https://www.sec.gov/Archives/edgar/data/<CIK without leading zeros>/<accessionNumber without dashes>/<primaryDocument>: the complete filing as HTML (inline XBRL). The filing's index page is the same folder plus <accessionNumber>-index.htm; cite that URL and the accession number in the References.
  4. Structured statement data (optional) — https://data.sec.gov/api/xbrl/companyfacts/CIK##########.json: every US-GAAP fact the company has tagged, by concept and period. Use it to confirm the figures read from Item 8.
  5. Full-text search (optional) — https://www.sec.gov/edgar/search/#/q=%22<phrase>%22&forms=10-K when a filing has to be located by phrase rather than ticker.

Send a User-Agent header that identifies the requester (the SEC requires it, e.g. Name email@example.com) and stay under 10 requests per second. Never substitute a summary site for the filing itself.

B. The host cannot fetch → Retrieval: pasted by user. Ask the user to paste the sections needed — or, if they have the InvestSkill repository, to run

node scripts/fetch-edgar.js <TICKER> --form 10-K        # → data/filings/<TICKER>/*.txt + .json (accession, dates, URL)
node scripts/fetch-fundamentals.js <TICKER>             # → data/fixtures/<TICKER>.md — reconciled statements from the XBRL facts

and paste the result. Both helpers are optional, keyless, and live outside the plugin.

C. Neither → show the ⚠️ Source unavailable warning, set Retrieval: model memory and Confidence: LOW, and digest only what the user provided.


How to Use

# Analyze by ticker (retrieves the latest 10-K via the EDGAR path above, or asks you to paste it)
/10k-digest AAPL

# Paste 10-K text directly
/10k-digest [paste 10-K sections here]

# Specify fiscal year
/10k-digest MSFT FY2024

# Output in Traditional Chinese
/10k-digest NVDA --lang zh-TW

# Save output to file
/10k-digest GOOGL --output googl-10k-digest.md

Output Document Structure

The skill produces a single markdown document with the following sections in order.


1. Cover Block

# [Company Name] ([TICKER]) — 10-K Digest
**Fiscal Year:** [FY20XX ending MM/DD/YYYY]
**Filed:** [Filing date]
**Source:** SEC EDGAR — [Direct URL to filing]
**Prepared:** [Today's date]
**Language:** English | 繁體中文

2. Abstract

A 3–5 sentence executive summary of the entire 10-K. Cover:

  • What the company does (core business in one sentence)
  • Year-over-year financial performance headline (revenue, profit, FCF)
  • The single most important strategic development disclosed in this filing
  • The most material risk disclosed
  • Overall filing tone (transparent / cautious / optimistic)

3. Section Summaries

Summarize each major 10-K section. For each section, write 2–4 sentences capturing the substance — not just restating headings.

3.1 Business Overview (Item 1)
  • Core products/services and revenue model
  • Key markets and customer segments
  • Significant changes to business described in this filing (new products, acquisitions, divestitures)
3.2 Risk Factors (Item 1A)
  • Total number of risk factors disclosed
  • Top 3 risks by disclosure prominence (listed first, given most space, or new this year)
  • Any risks added or removed vs. prior filing (state "N/A if prior not available")
  • One-line assessment: are risks standard boilerplate or specific and material?
3.3 Selected Financial Data / Properties (Items 1B–2)
  • Key quantitative highlights from Selected Financial Data if present
  • Material properties or facilities disclosed
3.4 Management's Discussion & Analysis — MD&A (Item 7)
  • Revenue: total and YoY growth; organic vs. inorganic split if stated
  • Gross margin and operating margin trend
  • Management's stated explanation for performance changes
  • Forward-looking statements: what management guided or projected
  • Liquidity summary: cash position, debt level, upcoming maturities
3.5 Financial Statements (Items 8)
  • Revenue, net income, EPS (GAAP) — current year vs. prior year
  • Operating cash flow and free cash flow (OCF minus capex)
  • Total assets, total debt, shareholders' equity
  • Auditor name and opinion type (unqualified / qualified / adverse)
  • Any restatements or material changes in accounting policy noted
3.6 Quantitative Disclosures about Market Risk (Item 7A)
  • Primary market risks disclosed (interest rate, FX, commodity)
  • Sensitivity disclosures if provided (e.g., "1% rate rise = $X impact")
3.7 Controls and Procedures (Item 9A)
  • Management's conclusion on effectiveness of internal controls
  • Any material weaknesses or significant deficiencies disclosed
3.8 Legal Proceedings (Item 3)
  • Active material litigation or regulatory matters
  • Estimated financial exposure if disclosed
3.9 Executive Compensation & Governance (Items 10–13, from Proxy if incorporated)
  • CEO/CFO compensation summary
  • Board composition notes (independence, tenure)
  • Any significant governance changes

4. Key Metrics Table

Provide a concise metrics table drawn directly from the financial statements. Use exact figures from the filing; do not estimate.

| Metric                    | Current FY      | Prior FY        | YoY Change |
|---------------------------|-----------------|-----------------|------------|
| Revenue                   | $X.XX B         | $X.XX B         | +X.X%      |
| Gross Profit              | $X.XX B         | $X.XX B         | +X.X%      |
| Gross Margin %            | XX.X%           | XX.X%           | +X bps     |
| Operating Income          | $X.XX B         | $X.XX B         |            |
| Operating Margin %        | XX.X%           | XX.X%           |            |
| Net Income (GAAP)         | $X.XX B         | $X.XX B         |            |
| EPS Diluted (GAAP)        | $X.XX           | $X.XX           |            |
| Operating Cash Flow       | $X.XX B         | $X.XX B         |            |
| Capital Expenditures      | $X.XX B         | $X.XX B         |            |
| Free Cash Flow            | $X.XX B         | $X.XX B         |            |
| Cash & Equivalents        | $X.XX B         | $X.XX B         |            |
| Total Debt                | $X.XX B         | $X.XX B         |            |
| Net Debt / (Net Cash)     | $X.XX B         | $X.XX B         |            |
| Shares Outstanding (dil.) | X.XX B          | X.XX B          |            |

Note below the table: data source (filing section and page number if available).


5. Highlights & Concerns

Two short lists that give the reader the "so what" without leaving the digest. Draw every item directly from the filing and cite the source Item.

Highlights — up to 5 of the strongest positives disclosed this year (e.g., record revenue, margin expansion, new products/segments, strong FCF conversion, new buyback authorization, market-share gains). Format: - **[Headline]** — [one sentence] *(Item X)*

Concerns — up to 5 of the most material negatives or warning signs (e.g., decelerating growth, margin compression, rising leverage, customer/supplier concentration, new risk factors, active litigation, going-concern language). Format: - **[Headline]** — [one sentence] *(Item X)*


6. Management Outlook & Forward Guidance

Capture every forward-looking commitment made in the filing. Present as a table.

| Guidance Item              | What Management Said                | Time Horizon | Source     |
|----------------------------|-------------------------------------|--------------|------------|
| Revenue growth             | [target / range / directional]      | [FY / multi] | Item 7     |
| Margin target              | [gross / operating target]          | [horizon]    | Item 7     |
| Capex / capacity           | [planned spend or expansion]        | [horizon]    | Item 7     |
| Capital allocation         | [dividend / buyback / debt / M&A]   | [horizon]    | Item 7/8   |
| Strategic priorities       | [product roadmap, new markets]      | [horizon]    | Item 1/7   |

Only include rows the filing actually addresses. If the 10-K gives no explicit guidance (common — 10-Ks are backward-looking), state that and summarize any qualitative forward statements instead.


7. Prior-Year Promise Tracker

An accountability audit: take the forward-looking statements, targets, and stated priorities from last year's 10-K (Item 7 MD&A) and score whether they materialized in this year's results. This is the single most differentiating part of the digest — it measures management credibility over time.

| Prior-Year Commitment (FY[N-1] 10-K)   | FY[N] Actual Result          | Verdict     | Notes            |
|-----------------------------------------|------------------------------|-------------|------------------|
| [what management said they would do]    | [what actually happened]     | ✅ Delivered | [brief context]  |
| [e.g., "expand operating margin"]       | [e.g., "margin +120 bps"]    | 🟡 Partial  | [brief context]  |
| [e.g., "launch product X in H2"]        | [e.g., "slipped to next FY"] | ❌ Missed    | [brief context]  |

Verdict legend: ✅ Delivered · 🟡 Partial · ❌ Missed · ⏳ Too early to judge.

Close with a one-line Management Credibility read: e.g., "4/5 prior commitments delivered — high follow-through." If the prior-year 10-K is unavailable, state N/A — prior filing not retrieved and skip the table.


8. Bull vs. Bear Synthesis

A lightweight face-off built strictly from evidence in this filing — not a full valuation call (defer to bear-case, stock-eval, and stock-valuation for that).

Bull case — 3–4 strongest points from this filing supporting the stock.

Bear case — 3–4 strongest points from this filing arguing against it.

Net read — one sentence weighing the two, consistent with the Investment Signal below.


9. Notable Disclosures

Bullet list of up to 8 items that are worth investor attention but don't fit neatly into section summaries:

  • New segment or business unit disclosures
  • Changes in accounting estimates or policies
  • Related party transactions
  • Auditor changes
  • Going concern language (if any)
  • New share repurchase authorizations or dividend changes
  • Material acquisitions or divestitures announced
  • Specific named customers or suppliers with concentration risk

10. Source References

List all sources cited in this digest. Format:

## References

1. **SEC EDGAR Filing** — [Company] 10-K, FY[XXXX]
   URL: https://www.sec.gov/cgi-bin/browse-edgar?...
   Filed: [Date] | Accession No.: [XXXXXXXXXX-XX-XXXXXX]

2. **Item 1 — Business** (pp. X–X of filing)
3. **Item 1A — Risk Factors** (pp. X–X of filing)
4. **Item 7 — MD&A** (pp. X–X of filing)
5. **Item 8 — Financial Statements** (pp. X–X of filing)
6. **Item 9A — Controls and Procedures** (pp. X–X of filing)
7. **Exhibit 13 / Annual Report to Shareholders** (if incorporated by reference)
8. [Any additional sources used, e.g. earnings press release, investor presentation]

If page numbers are unavailable (e.g., pasted text), reference by Item number and section heading.


Language Mode

When --lang zh-TW is specified, output the entire document in Traditional Chinese (繁體中文), including:

  • All section headings translated
  • All narrative summaries written in Traditional Chinese
  • All tables with Chinese column headers
  • Cover block language field: 語言:繁體中文
  • Keep ticker symbols, financial figures, and SEC item numbers in their original form

Suggested Traditional Chinese headings for the interpretation sections: 亮點與隱憂 (Highlights & Concerns), 管理層展望 (Management Outlook & Forward Guidance), 上年度承諾兌現 (Prior-Year Promise Tracker), 多空對決 (Bull vs. Bear Synthesis).

English mode is the default when no --lang flag is given.


File Output

When --output <filename> is specified:

  • Save the digest as a markdown file at the specified path
  • If no path separator is present, save in the current directory
  • Confirm the file path after saving: ✓ Digest saved to: <path>
  • Default filename if --output flag is present without a name: <TICKER>-10k-digest.md

Thesis Invalidation

After delivering the digest, note what would make this document stale:

Re-run this digest when:

  • Company files next 10-K or 10-K/A amendment
  • Material 8-K filed disclosing a restatement or auditor change
  • 12 months have elapsed since the filing date

In the output, the box below is followed immediately by this machine-readable footer — the same values as the box, filled in (no placeholders, no comments), so tools can read the result without parsing the box:

{"investskill": "1", "skill": "10k-digest", "ticker": "<TICKER, or null for a portfolio / market-wide view>", "as_of": "<YYYY-MM-DD from Data & Sources>", "signal": "<BULLISH | NEUTRAL | BEARISH>", "confidence": "<HIGH | MEDIUM | LOW>", "horizon": "<SHORT-TERM | MEDIUM-TERM | LONG-TERM>", "score": 0.0, "action": "<BUY | HOLD | SELL>", "conviction": "<STRONG | MODERATE | WEAK>", "retrieval": "<pasted by user | web/tool retrieval | model memory>", "sources": ["<each primary source, dated>"]}
╔══════════════════════════════════════════════╗
║              INVESTMENT SIGNAL               ║
╠══════════════════════════════════════════════╣
║ Signal:      BULLISH / NEUTRAL / BEARISH     ║
║ Confidence:  HIGH / MEDIUM / LOW             ║
║ Horizon:     SHORT / MEDIUM / LONG-TERM      ║
║ Score:       X.X / 10                        ║
╠══════════════════════════════════════════════╣
║ Action:      BUY / HOLD / SELL               ║
║ Conviction:  STRONG / MODERATE / WEAK        ║
╚══════════════════════════════════════════════╝

Score Guide: 8.0–10.0 Strongly Bullish | 6.0–7.9 Moderately Bullish | 4.0–5.9 Neutral | 2.0–3.9 Moderately Bearish | 0.0–1.9 Strongly Bearish Confidence: HIGH (strong data, clear signals) | MEDIUM (mixed signals) | LOW (limited data, conflicting signals) Horizon: SHORT-TERM (1 week–3 months) | MEDIUM-TERM (3 months–1 year) | LONG-TERM (1+ years)

Disclaimer: Educational analysis only. Not financial advice.

Basket Report

⚠️ Data Verification — Do This Before Any Analysis

Before running any analysis, always retrieve the latest market data for every ticker in the basket:

  1. Fetch current price — use web search or ask the user for the live price, 52-week range, and market cap. Never assume a price from training data.
  2. Confirm key figures — for every ticker, on the same as-of date: price, market cap, the last four fiscal years and five quarters of statements, shares outstanding, consensus estimates and the next earnings date. Gather them once, save them as the basket snapshot (Phase 1), and run every module from that snapshot — a basket in which two pages quote different prices for the same stock has failed before any analysis starts.
  3. State your data source — fill in the Data & Sources header (next section) so the origin, as-of date, retrieval path, and confidence of every figure are explicit at the top of the output.
  4. Flag stale data explicitly — if live data is unavailable, display this warning before proceeding:

⚠️ Live data unavailable. The following analysis uses training-data estimates which may be significantly out of date. Verify all prices and metrics before making any decisions.

Never silently substitute training-data estimates for current prices. When in doubt, ask the user to paste the latest quote.


📋 Data & Sources Header — Open Every Output With It

The first thing in the output is this provenance block, filled in — never left as placeholders. It is the standard documented on the Data & Accuracy page and the first thing result-validator looks for:

Data & Sources
  As of:      <date the figures represent, e.g. 2026-06-30>
  Source:     <primary docs — SEC EDGAR 10-K/10-Q, company IR, FRED, exchange data …>
  Retrieval:  <pasted by user | web/tool retrieval | model memory>
  Confidence: <HIGH | MEDIUM | LOW>
  • Retrieval: model memory must be paired with Confidence: LOW — memory is a placeholder until confirmed against a primary source.
  • Mixed sources: list each with its own as-of date rather than blending them.
  • Sources disagree on the same figure (a closing price, a deal term): prefer the primary source (SEC filing, company release, exchange data) over an aggregator, then the most recent dated one. Show both values with their sources and dates, say which you used and why, and never average them.
  • Data the user pasted is reported as pasted by user; do not upgrade its confidence beyond what the user's own source supports.

🌐 Output Language — --lang zh-TW

English is the default. With --lang zh-TW — or when the user writes in Traditional Chinese and asks for a Chinese answer — write the whole analysis in Traditional Chinese (繁體中文, Taiwan usage), keeping it machine-checkable:

  • Translate every heading, narrative paragraph, table header and bullet. Give each metric's English name in brackets the first time it appears — 自由現金流(Free Cash Flow, FCF) — so the reader can match it to the glossary and to English analyses.
  • Keep in English tickers, figures and units, SEC form and item names (10-K Item 7, Form 4), the four Data & Sources field labels (As of, Source, Retrieval, Confidence — the Retrieval value may be written in Chinese, e.g. 使用者提供), and the Investment Signal box and JSON footer exactly as specified — field names and the BULLISH / NEUTRAL / BEARISH, HIGH / MEDIUM / LOW, BUY / HOLD / SELL, STRONG / MODERATE / WEAK values. Tools parse them.
  • Title the invalidation section 論點失效條件(Thesis Invalidation), and close with the disclaimer in Chinese: 免責聲明: 僅供教育用途,非投資建議。
  • Use Taiwan financial terms: 殖利率、本益比、營業利益率、自由現金流、聯準會、選擇權、買權/賣權、除息日、預扣稅.

Overview

full-report answers "tell me everything about this one stock". compare answers "which of these should I own?" and ends in one Markdown table. Neither produces what a person researching a small group of names actually wants to keep: one deliverable in which every stock has its own report, every report quotes the same prices, and a front page says what the group has in common and which one to pick. Running full-report four times gives four files written on four sets of assumptions, with nothing joining them.

This skill is an orchestrator, not a new analysis. It runs existing frameworks over a basket of 2–5 tickers from one shared data snapshot and assembles the results into a small linked site: a hub page, one report per ticker, and a value-chain page when the names share one. It adds no scoring of its own — every number on the hub comes from compare or from a per-ticker full-report, and the closing signal is the pick's.

SkillQuestion it answersOutput
stock-screenerWhich of these 20–50 names deserve a closer look?One ranked table
compareWhich of these 2–5 should I own, and why?One table and an argued pick (Markdown)
full-reportWhat is the whole picture on this one stock?One HTML report
basket-reportWhat is the whole picture on these 2–5 stocks, together?A hub page + one report per ticker, linked, on one snapshot

It is the most expensive skill in the plugin. At the default depth it runs 10 modules per ticker, so a four-stock basket is 40 module runs before the hub. Say so before starting, and offer --depth quick (5 per ticker) when the user has not asked for depth.


1. Inputs

InputRequiredDefault if unstated
Tickers✅ 2–5One ticker → stop and suggest full-report. More than five → run stock-screener first, carry its top five forward, and keep its leaderboard as an appendix on the hub. An ETF in the list → analyse it with etf-analysis separately; it is not a basket member
The questionrecommendedIf the user gives none, propose one from the data (see Phase 2) and state that it was proposed
--depthoptionalstandard (10 modules per ticker). quick (5) is allowed. comprehensive is not offered here — run full-report --depth comprehensive on the pick afterwards
--skip a,b,…optionalPassed to every per-ticker full-report; the same modules are skipped for every ticker so the reports stay comparable
Objective and horizonoptionalPassed to compare (default: total return over 3–5 years, LONG-TERM)
Existing holdingsoptionalPassed to compare for the diversification check
--langoptionalen; zh-TW supported (section above)
--outputoptionaloutput/

2. Framework

The five rules

These hold for every phase. They are what separates a basket report from several reports stapled together.

  1. One snapshot, one as-of date. All data for all tickers is gathered once, before any analysis, and saved. Every module of every report reads that snapshot and nothing else. No module re-fetches a price mid-run: two pages that quote different prices for the same stock are a defect, whichever is newer.
  2. Show the input beside the score. Every score on the hub and in each scorecard appears with the raw value it came from and the weight applied (EV/EBIT 18× → 6.9, Valuation 6.9 × 25% = 1.73). A score without its input cannot be checked.
  3. Missing is not neutral. A metric, module or dimension that cannot be computed for one ticker is written as not assessed, with the reason, and the weights that were actually used are shown. It is never filled with 5.0, and never copied from another ticker.
  4. Defects stay visible. When the data has a problem — statements in one currency and the price in another, a listing too recent for moving averages, a vendor field that disagrees with the filing — keep the raw value, flag it where it appears, and list it in the hub's defect ledger. Do not silently repair it and do not drop the ticker.
  5. A refusal is a result. If a ticker's data cannot support a valuation or a target, its report says so and gives none. If compare finds the group not comparable, the hub says so and ranks within subgroups. If no ticker is attractive, the pick is "none of these". Each of these is a finished basket report.

Phase 1 — The snapshot

Gather, for every ticker on the same as-of date: price, 52-week range and market cap; the last four fiscal years and five quarters of income statement, balance sheet and cash flow; shares outstanding; consensus estimates and the next earnings date; insider transactions and top institutional holders; short interest; one year of price history. Record the source and retrieval path of each group of figures.

Save it as snapshot.md in the output folder, one section per ticker, before any analysis. From here on the snapshot is the only input. If a figure is missing for one ticker, write missing — <reason> in its place rather than leaving the row out.

Then open the output with one Data & Sources header for the whole basket. If a ticker's figures could only be obtained for a different date, list it on its own line with that date, and cap the basket's Confidence at MEDIUM.

Phase 2 — The question

A basket needs a reason to be a basket. State it in one sentence at the top of the hub, as a question the reports then answer: "Three memory names fell together on one day — is this the cycle turning, or a liquidity event?" Use the user's question if they gave one. Otherwise propose the question the snapshot itself raises — a shared move, a shared driver, a valuation spread between similar businesses — and label it as proposed.

Also state, in one line each, why each ticker is in the basket: its role (the pure play, the supplier, the incumbent, the control).

Phase 3 — One report per ticker

For each ticker run full-report at the chosen depth, with the snapshot as its data and the same --skip list for all. Follow its phase order — business quality, then valuation, then market signals, technical setup and risk — because earlier phases feed later ones.

Finish and save each ticker's report before starting the next. A basket does not fit in working memory; the saved file is the hand-off. Each report keeps full-report's own contract: its Data & Sources header (identical to the basket's), its scorecard, its Thesis Invalidation, its own signal box and JSON footer.

Phase 4 — The comparison and the pick

Run compare on the basket, taking its inputs from the snapshot and its moat, reverse-DCF and catalyst inputs from the Phase 3 reports. It supplies the head-to-head table, the argued pick, the flip condition, the diversification answer and the "same business, different price" check.

Two rankings now exist: compare's composite (absolute anchors, seven dimensions) and each ticker's full-report composite (its own modules). Show both on the hub, labelled. If they disagree on the order of the top two by more than 0.5, say so, explain which dimension causes it, and argue the pick on the user's objective — do not quietly report one.

Phase 5 — The value chain (when there is one)

If two or more tickers sit in the same supply or value chain, run industry-map once for that chain and mark each basket member's position, the chokepoints and the margin pools. If the tickers share no chain, write "no shared value chain — industry-map not applicable" on the hub and skip the page.

Phase 6 — Audit before assembly

  1. Run result-validator on each per-ticker composite. A report scoring LOW carries that warning at the top of its page and beside its row on the hub.
  2. Check the basket for agreement: each ticker's price, market cap and as-of date are identical on every page that quotes them; every figure repeated between a ticker's report and the hub matches; the hub's composite for each ticker equals the one in that ticker's scorecard. Fix a disagreement at its source and re-render; do not edit one page to match another.
  3. Write the defect ledger (Rule 4): each data problem found, the ticker, the pages affected, and what was done about it.

Phase 7 — The hub and the signal

Assemble the hub page last, from the saved reports. Then close with the signal block. The box expresses the view on the pick, on an absolute basis:

  • The pick is the one argued in Phase 4.
  • The Score is the pick's full-report composite from Phase 3, so it equals the Score in the pick's own report.
  • Conviction cannot be STRONG when the two rankings disagreed in Phase 4, or when the pick's result-validator confidence is LOW.
  • If the pick is "none of these", the box reports the highest composite in the basket with Action: HOLD or SELL by the Score Guide, and the line above the box says so.

3. Output Format

Files, all in one folder so the links between them are relative:

output/basket_<slug>_<YYYY-MM-DD>/
├── index.html        # the hub
├── <ticker>.html     # one per ticker, from full-report
├── chain.html        # industry-map, only if Phase 5 ran
└── snapshot.md       # the shared data every page was built from

<slug> is the tickers in lower case joined by hyphens (mu-mrvl-sndl). Every page is a standalone HTML file that links to the others with relative paths and carries the same navigation: hub · each ticker · chain.

The hub, in order:

  1. Data & Sources header for the basket
  2. The question — one sentence, and whether the user or the skill posed it (Phase 2)
  3. The basket — a row per ticker: role, price, composite, signal, result-validator confidence, link to its report
  4. The answer — three sentences: the pick, the deciding dimension, the flip condition
  5. Head-to-head table — from compare, every cell raw → sub-score, with both composites per ticker (Phase 4)
  6. What they share — the common driver, pairwise correlation, and the own-one / own-both answer
  7. Value chain — a summary and link, or "not applicable" (Phase 5)
  8. Coverage — which modules ran for every ticker, which were skipped by request, which were not assessed and why
  9. Defect ledger (Phase 6)
  10. How this was built — the snapshot date, the depth, the module count, and the order of the phases
  11. Thesis Invalidation and the Investment Signal block — the line just above the box names the pick: "Signal for the pick: ZEPH"

In chat, reply with the hub's sections 1–4, the list of files written, and the signal block — not the full reports.


Example

User: Basket report on ZEPH QRTZ MNTL — which software name do I keep for five years?

[ZEPH is the plugin's fictional ticker; QRTZ and MNTL are hypothetical peers invented for
this example. All figures are illustrative.]

Before starting: 3 tickers × 10 modules = 30 module runs at standard depth, plus compare
and the hub. (--depth quick would be 15.) Proceeding at standard.

Snapshot saved: output/basket_zeph-qrtz-mntl_2026-11-14/snapshot.md — all three as of the
2026-11-13 close, pasted by the user. MNTL short interest: missing — not in the pasted data.

Question (yours): which of three profitable software companies to hold for five years.
Roles: ZEPH — the cheap one · QRTZ — the best business · MNTL — the fastest grower.

Reports written: zeph.html (composite 6.8) · qrtz.html (6.4) · mntl.html (5.7 — short-interest
not assessed, Risk Profile re-weighted across the modules that ran).

Pick: ZEPH. compare ranks ZEPH 7.0 / QRTZ 6.3 / MNTL 5.9; the full-report composites agree
on the order. Deciding dimension: valuation (EV/EBIT 18× → 6.9 against QRTZ 29× → 3.4) at
near-equal quality. Flip: QRTZ at the price where its composite reaches 7.0, or ZEPH's top
customer above 25% of revenue. ZEPH/QRTZ weekly correlation 0.83 → own one, not both.

Value chain: all three sell to enterprise IT budgets but none supplies another — no shared
chain, industry-map not applicable.

Defect ledger: 1 — MNTL reports stock compensation outside "adjusted" EBIT; all three are
shown after SBC, MNTL's adjusted figure kept beside it and flagged on mntl.html.

Signal for the pick: ZEPH — Score 6.8 (its full-report composite) → BULLISH, MODERATE,
LONG-TERM, Confidence MEDIUM (user-pasted data, one field missing).
Files: index.html, zeph.html, qrtz.html, mntl.html, snapshot.md

Notes

  • Two to five tickers. Fewer is a full-report; more stops being a report anyone reads. Use stock-screener to get down to five.
  • Standard depth is the ceiling. The basket's job is breadth on one snapshot. Depth belongs to the pick: follow with full-report --depth comprehensive, then thesis-tracker.
  • Skip the same modules for every ticker. A basket in which one report has options-analysis and another does not is not comparing like with like. A module that cannot run for one ticker is "not assessed" there, not skipped everywhere.
  • The snapshot is a file the user can read. Offer it; a reader who doubts a number should be able to find where it came from.
  • This skill fetches nothing by itself. As everywhere in InvestSkill, the data comes from the user or from the host assistant's own tools; the rule here is only that it is gathered once.
  • Before: stock-screener to shortlist. After: full-report --depth comprehensive and bear-case on the pick; thesis-tracker to save it; position-ladder for sizing; risk-stress-test if more than one is owned.

Thesis Invalidation

After delivering the analysis signal, specify what would reverse it:

If signal is BULLISH (the pick is a buy) — thesis breaks if:

  • The flip condition on the hub is met — the runner-up reaches its flip price, or the pick's deciding KPI crosses the stated level
  • The pick's own report breaks: any trigger in that ticker's full-report Thesis Invalidation fires
  • An item in the defect ledger turns out to explain the pick's advantage (the cheap stock was cheap because of the data problem, not despite it)

If signal is BEARISH (none of the basket is attractive) — thesis breaks if:

  • The shared driver named under "What they share" turns favourable in reported numbers, not forecasts
  • A basket-wide price decline lifts the best valuation sub-score by 3 or more points with business quality unchanged
  • A "not assessed" module that was holding a composite down can now be run and scores above 6.0

Re-run this analysis when:

  • Next earnings release of any ticker in the basket (take a new snapshot for all of them, not just that one)
  • Price moves ±15% from the snapshot level for any ticker
  • 60 days have elapsed
  • Material news event (acquisition, leadership change, regulatory decision)

Standard Signal Output

The box expresses the view on the pick only (Phase 7). Name the pick in the line directly above the box ("Signal for the pick: "); each other ticker's signal is in its own report and in the hub's basket table.

All analysis concludes with this standardized block:

In the output, the box below is followed immediately by this machine-readable footer — the same values as the box, filled in (no placeholders, no comments), so tools can read the result without parsing the box:

{"investskill": "1", "skill": "basket-report", "ticker": "<TICKER, or null for a portfolio / market-wide view>", "as_of": "<YYYY-MM-DD from Data & Sources>", "signal": "<BULLISH | NEUTRAL | BEARISH>", "confidence": "<HIGH | MEDIUM | LOW>", "horizon": "<SHORT-TERM | MEDIUM-TERM | LONG-TERM>", "score": 0.0, "action": "<BUY | HOLD | SELL>", "conviction": "<STRONG | MODERATE | WEAK>", "retrieval": "<pasted by user | web/tool retrieval | model memory>", "sources": ["<each primary source, dated>"]}
╔══════════════════════════════════════════════╗
║              INVESTMENT SIGNAL               ║
╠══════════════════════════════════════════════╣
║ Signal:      BULLISH / NEUTRAL / BEARISH     ║
║ Confidence:  HIGH / MEDIUM / LOW             ║
║ Horizon:     SHORT / MEDIUM / LONG-TERM      ║
║ Score:       X.X / 10                        ║
╠══════════════════════════════════════════════╣
║ Action:      BUY / HOLD / SELL               ║
║ Conviction:  STRONG / MODERATE / WEAK        ║
╚══════════════════════════════════════════════╝

Score Guide: 8.0–10.0 Strongly Bullish | 6.0–7.9 Moderately Bullish | 4.0–5.9 Neutral | 2.0–3.9 Moderately Bearish | 0.0–1.9 Strongly Bearish Confidence: HIGH (strong data, clear signals) | MEDIUM (mixed signals) | LOW (limited data, conflicting signals) Horizon: SHORT-TERM (1 week–3 months) | MEDIUM-TERM (3 months–1 year) | LONG-TERM (1+ years)

Note: The Score in the box is the pick's full-report composite, so it equals the Score in that ticker's own report; compare's composites are shown beside it in the head-to-head table.

Disclaimer: Educational analysis only. Not financial advice.

Bear Case (Bearish Investor)

⚠️ Data Verification — Do This Before Any Analysis

Before running any analysis, always retrieve the latest market data for the ticker:

  1. Fetch current price — use web search or ask the user for the live price, 52-week range, and market cap. Never assume a price from training data.
  2. Confirm key figures — the latest quarter's revenue, margins and guidance, the valuation multiples the bull case leans on (EV/EBIT, P/FCF, P/S), net debt, share count trend, and short interest — each with its filing or quote date.
  3. State your data source — fill in the Data & Sources header (next section) so the origin, as-of date, retrieval path, and confidence of every figure are explicit at the top of the output.
  4. Flag stale data explicitly — if live data is unavailable, display this warning before proceeding:

⚠️ Live data unavailable. The following analysis uses training-data estimates which may be significantly out of date. Verify all prices and metrics before making any decisions.

Never silently substitute training-data estimates for current prices. When in doubt, ask the user to paste the latest quote.


📋 Data & Sources Header — Open Every Output With It

The first thing in the output is this provenance block, filled in — never left as placeholders. It is the standard documented on the Data & Accuracy page and the first thing result-validator looks for:

Data & Sources
  As of:      <date the figures represent, e.g. 2026-06-30>
  Source:     <primary docs — SEC EDGAR 10-K/10-Q, company IR, FRED, exchange data …>
  Retrieval:  <pasted by user | web/tool retrieval | model memory>
  Confidence: <HIGH | MEDIUM | LOW>
  • Retrieval: model memory must be paired with Confidence: LOW — memory is a placeholder until confirmed against a primary source.
  • Mixed sources: list each with its own as-of date rather than blending them.
  • Sources disagree on the same figure (a closing price, a deal term): prefer the primary source (SEC filing, company release, exchange data) over an aggregator, then the most recent dated one. Show both values with their sources and dates, say which you used and why, and never average them.
  • Data the user pasted is reported as pasted by user; do not upgrade its confidence beyond what the user's own source supports.

🌐 Output Language — --lang zh-TW

English is the default. With --lang zh-TW — or when the user writes in Traditional Chinese and asks for a Chinese answer — write the whole analysis in Traditional Chinese (繁體中文, Taiwan usage), keeping it machine-checkable:

  • Translate every heading, narrative paragraph, table header and bullet. Give each metric's English name in brackets the first time it appears — 自由現金流(Free Cash Flow, FCF) — so the reader can match it to the glossary and to English analyses.
  • Keep in English tickers, figures and units, SEC form and item names (10-K Item 7, Form 4), the four Data & Sources field labels (As of, Source, Retrieval, Confidence — the Retrieval value may be written in Chinese, e.g. 使用者提供), and the Investment Signal box and JSON footer exactly as specified — field names and the BULLISH / NEUTRAL / BEARISH, HIGH / MEDIUM / LOW, BUY / HOLD / SELL, STRONG / MODERATE / WEAK values. Tools parse them.
  • Title the invalidation section 論點失效條件(Thesis Invalidation), and close with the disclaimer in Chinese: 免責聲明: 僅供教育用途,非投資建議。
  • Use Taiwan financial terms: 殖利率、本益比、營業利益率、自由現金流、聯準會、選擇權、買權/賣權、除息日、預扣稅.

⚠️ Read This First — This Analysis Is Intentionally One-Sided

You are a skeptical short-seller and professional bear. Your single mandate is to construct the strongest, most intellectually honest case for why this stock should NOT be held — and, in the extreme, why it should be sold or shorted.

This is a red-team / devil's-advocate tool by design. It is deliberately biased to the downside. Its value comes from being one-sided: it forces the counterevidence to the surface so the user can stress-test a bullish thesis and see the stock from the inverse direction. It is not a balanced call and must never be presented as one.

  • Always open the output with: "This is a deliberately one-sided bear case. Pair it with /stock-eval (or a bull-case analysis) for a balanced view."
  • Argue the bear side with conviction, but never fabricate. Every claim must be grounded in real data, a real risk, or an explicitly labeled assumption/hypothesis.
  • Steelman the bear thesis, then be honest about what would break it (the "Thesis-Killers" section is mandatory).

Construct a rigorous short thesis for a US-listed stock: identify overvaluation, deteriorating fundamentals, accounting and quality red flags, competitive and secular threats, weak management and capital allocation, and concrete downside catalysts — then quantify the downside and define what would prove the bear wrong.

Analysis Framework

1. Valuation Stretch — "The Price Already Assumes Too Much"

The most common bear case: the market is paying for perfection.

  • Multiple vs. history and peers — Is P/E, P/S, EV/EBITDA, EV/Sales, or P/FCF elevated vs. the stock's own 5-year range and vs. sector peers? By how many turns?
  • Expectations embedded in the price — Reverse-DCF: what growth rate / margin does today's price require? Is that plausible, or does it demand flawless execution for a decade?
  • Priced for perfection — Any deceleration, miss, or guide-down that the market has not discounted.
  • PEG and growth-adjusted value — Is the growth premium justified by durable growth, or by a one-time / cyclical surge?
  • Quality of the multiple — Is the "cheap" multiple a value trap (declining business) rather than a bargain?

2. Deteriorating Fundamentals — "The Business Is Getting Worse"

  • Revenue trajectory — Decelerating growth, tougher comps, saturation, or outright declines.
  • Margin compression — Gross/operating/net margin trend breaking down; input-cost, pricing, or mix pressure.
  • Cash flow quality — FCF diverging from net income; rising capex intensity; negative or shrinking FCF.
  • Balance sheet stress — Rising leverage, near-term maturities, covenant risk, interest-coverage deterioration, rising net debt / EBITDA.
  • Working capital red flags — Inventory building faster than sales; receivables (DSO) rising (channel stuffing risk).
  • Returns on capital — Falling ROIC/ROE, especially ROIC below WACC (value-destroying growth).

3. Accounting & Earnings-Quality Red Flags — "The Numbers May Not Be Real"

Apply forensic-accounting skepticism (Hindenburg / Muddy Waters / Beneish-style lenses):

  • Aggressive revenue recognition — Bill-and-hold, percentage-of-completion abuse, related-party revenue.
  • Non-GAAP gaming — Large or growing gap between GAAP and "adjusted" earnings; recurring "one-time" charges; stock-based comp excluded from adjusted metrics.
  • Cash vs. accrual divergence — Net income up while operating cash flow lags = low earnings quality.
  • Buyback-masked dilution — Buybacks offsetting heavy SBC rather than reducing share count; per-share metrics flattered by financial engineering.
  • DSO / DIO trends — Rising days-sales-outstanding or days-inventory (Beneish DSRI/SGI signals).
  • Auditor / disclosure issues — Auditor changes, restatements, late filings, material-weakness disclosures, CFO turnover.

4. Competitive & Secular Threats — "The Moat Is Eroding"

  • Moat erosion — Loss of pricing power, share loss to rivals, commoditization.
  • Disruption / obsolescence — New entrant, technology shift, or substitute product structurally impairing the model.
  • Secular decline — Industry in structural (not cyclical) contraction; TAM shrinking.
  • Customer & supplier concentration — Dependence on a few customers/suppliers; key-customer churn risk.
  • Regulatory / legal overhang — Antitrust, litigation, tariffs, pending investigations, product-liability exposure.

5. Management & Capital Allocation — "Stewardship Is Poor"

  • Capital-allocation track record — Value-destructive M&A, buybacks at peak prices, dividends funded by debt.
  • Insider behavior — Heavy insider selling, option-heavy comp, misaligned incentives.
  • Governance red flags — Dual-class control, related-party transactions, board entrenchment, aggressive guidance history.
  • Credibility — Repeated guide-downs, promotional tone, or a widening gap between narrative and results.

6. Downside Catalysts & Timeline — "What Breaks It, and When"

A bear thesis without a catalyst is just an opinion. Identify what forces repricing:

  • Near-term (0–6 mo) — Next earnings/guide-down, expiring lockup, debt maturity, product cycle miss, key data point.
  • Medium-term (6–18 mo) — Margin normalization, competitive launch, regulatory decision, refinancing at higher rates.
  • Structural — Multiple compression as growth fades; index removal; secular demand roll-over.
  • For each, note the trigger, likely magnitude, and timing.

7. Downside Quantification — "How Far Can It Fall"

  • Bear price target — Apply a de-rated multiple to conservative (bear-case) estimates. Show the math.
  • Downside scenario tree — Base-bear vs. severe-bear (e.g., recession + multiple compression) with rough probabilities.
  • Risk/reward from here — Downside-to-target vs. upside-if-wrong; is the asymmetry favorable to a short / to avoiding the name?
  • Fundamental floor — Book value, net cash, or asset value that limits downside (honest bears state the floor).

8. Thesis-Killers — "What Would Prove the Bear Wrong" (Mandatory)

Intellectual honesty is what separates a credible bear from a permabear. Explicitly list:

  • The 3–5 developments that would invalidate the short thesis (e.g., margin re-acceleration, successful new product, deleveraging, activist/takeover interest).
  • The strongest bull counterarguments and why the bear still disagrees (or concedes).
  • The biggest risk to being short: valuation support, squeeze potential (cross-reference /short-interest), takeout risk, or a fundamental floor.

Input Formats

Format 1: Single-Stock Bear Case

User: /bear-case TSLA

Claude builds the full short thesis and quantifies downside for TSLA.

Format 2: Counter a Bullish Thesis (Inverse Mode)

User: /bear-case NVDA — here is my bull thesis: [paste]

Claude attacks each pillar of the bull thesis and surfaces the counterevidence.

Format 3: Red-Team a Prior Analysis

User: /bear-case — challenge this stock-eval output: [paste]

Claude takes the opposing side of the prior (likely balanced or bullish) analysis.

Output

Provide a structured bear-case report:

1. One-Sided-Disclosure Banner

⚠️ This is a deliberately one-sided bear case built to surface counterevidence. Pair it with /stock-eval or a bull analysis for a balanced view.

2. Bear Thesis in Three Sentences

The elevator pitch for why the stock is a bad hold.

3. Bear Case Strength Score (0–10)

Pillar                              Weight   Score
Valuation stretch                    0–2      X.X
Deteriorating fundamentals           0–2      X.X
Accounting / earnings-quality flags  0–2      X.X
Competitive & secular threats        0–2      X.X
Management & capital allocation      0–1      X.X
Catalyst clarity & timing            0–1      X.X
                          BEAR CASE STRENGTH:  X.X / 10
Bear Case Strength   Interpretation
0.0–2.0              Weak — few credible negatives; bull case likely intact
2.1–4.0              Modest — some concerns, not thesis-breaking
4.1–6.0              Moderate — real red flags; reduce/hedge worth considering
6.1–8.0              Strong — multiple independent negatives; avoid / short candidate
8.1–10.0            Severe — deep impairment or fraud-risk signals; high-conviction bear

4. Detailed Findings

Cover pillars 1–6 above, each with the evidence and its severity (pillar 7 feeds §5, pillar 8 is §6).

5. Downside Target & Risk/Reward

Bear price target with the math, scenario tree, and asymmetry assessment.

6. Thesis-Killers

The mandatory list of what would prove the bear wrong.

Thesis Invalidation

After delivering the analysis signal, specify what would reverse it:

If signal is BULLISH — thesis breaks if:

  • Fundamentals deteriorate: margin compression, decelerating revenue, or a guide-down
  • [One or two more triggers drawn from this analysis's own drivers, each with a threshold]

If signal is BEARISH — thesis breaks if:

  • Margin or revenue growth re-accelerates and beats consensus with a guidance raise
  • A thesis-killer fires: successful new product, deleveraging, activist/takeover interest, or a short squeeze
  • Valuation compresses to a level that already prices in the bear case (downside exhausted)

Re-run this analysis when:

  • Next earnings release
  • Price moves ±15% from current level
  • 60 days have elapsed
  • Material news event (acquisition, leadership change, regulatory decision)

Signal Output

The Bear Case Strength Score maps to the standard signal as follows: a stronger bear case (higher score) means a more BEARISH signal. A strong/severe bear case → BEARISH / SELL; a weak bear case → the bear failed to make its point, leaning NEUTRAL-to-constructive. Because this skill argues one side, state the mapping explicitly so the user reads the block correctly:

Box Score = min(10 − Bear Case Strength, 6.5)

Strength 8.0 → Score 2.0 (BEARISH) · Strength 5.0 → Score 5.0 (NEUTRAL) · Strength 1.5 → Score 6.5 (BULLISH, WEAK). The cap is deliberate: a bear case that fails to land means the bull thesis survived scrutiny, not that it is strong — run stock-eval or stock-valuation for the positive case. Print both numbers: "Bear Case Strength 7.2 → Score 2.8".

All analysis concludes with this standardized block:

In the output, the box below is followed immediately by this machine-readable footer — the same values as the box, filled in (no placeholders, no comments), so tools can read the result without parsing the box:

{"investskill": "1", "skill": "bear-case", "ticker": "<TICKER, or null for a portfolio / market-wide view>", "as_of": "<YYYY-MM-DD from Data & Sources>", "signal": "<BULLISH | NEUTRAL | BEARISH>", "confidence": "<HIGH | MEDIUM | LOW>", "horizon": "<SHORT-TERM | MEDIUM-TERM | LONG-TERM>", "score": 0.0, "action": "<BUY | HOLD | SELL>", "conviction": "<STRONG | MODERATE | WEAK>", "retrieval": "<pasted by user | web/tool retrieval | model memory>", "sources": ["<each primary source, dated>"]}
╔══════════════════════════════════════════════╗
║              INVESTMENT SIGNAL               ║
╠══════════════════════════════════════════════╣
║ Signal:      BULLISH / NEUTRAL / BEARISH     ║
║ Confidence:  HIGH / MEDIUM / LOW             ║
║ Horizon:     SHORT / MEDIUM / LONG-TERM      ║
║ Score:       X.X / 10                        ║
╠══════════════════════════════════════════════╣
║ Action:      BUY / HOLD / SELL               ║
║ Conviction:  STRONG / MODERATE / WEAK        ║
╚══════════════════════════════════════════════╝

Score Guide: 8.0–10.0 Strongly Bullish | 6.0–7.9 Moderately Bullish | 4.0–5.9 Neutral | 2.0–3.9 Moderately Bearish | 0.0–1.9 Strongly Bearish Confidence: HIGH (strong data, clear signals) | MEDIUM (mixed signals) | LOW (limited data, conflicting signals) Horizon: SHORT-TERM (1 week–3 months) | MEDIUM-TERM (3 months–1 year) | LONG-TERM (1+ years)

Note: The Score above uses the standard bullish scale for cross-skill comparability. A strong bear case produces a LOW score (bearish). Because this analysis is deliberately one-sided, always pair it with /stock-eval before acting.

Disclaimer: Educational analysis only. Not financial advice.

Catalyst Calendar

⚠️ Data Verification — Do This Before Any Analysis

Before running any analysis, always retrieve the latest market data for the ticker:

  1. Fetch current price — use web search or ask the user for the live price, 52-week range, and market cap. Never assume a price from training data.
  2. Confirm key figures — the confirmed next earnings date (company IR, not an estimate), ex-dividend and record dates, the FOMC/CPI calendar, lock-up and index-rebalance dates, and any regulatory or court dates — each with its source.
  3. State your data source — fill in the Data & Sources header (next section) so the origin, as-of date, retrieval path, and confidence of every figure are explicit at the top of the output.
  4. Flag stale data explicitly — if live data is unavailable, display this warning before proceeding:

⚠️ Live data unavailable. The following analysis uses training-data estimates which may be significantly out of date. Verify all prices and metrics before making any decisions.

Never silently substitute training-data estimates for current prices. When in doubt, ask the user to paste the latest quote.


📋 Data & Sources Header — Open Every Output With It

The first thing in the output is this provenance block, filled in — never left as placeholders. It is the standard documented on the Data & Accuracy page and the first thing result-validator looks for:

Data & Sources
  As of:      <date the figures represent, e.g. 2026-06-30>
  Source:     <primary docs — SEC EDGAR 10-K/10-Q, company IR, FRED, exchange data …>
  Retrieval:  <pasted by user | web/tool retrieval | model memory>
  Confidence: <HIGH | MEDIUM | LOW>
  • Retrieval: model memory must be paired with Confidence: LOW — memory is a placeholder until confirmed against a primary source.
  • Mixed sources: list each with its own as-of date rather than blending them.
  • Sources disagree on the same figure (a closing price, a deal term): prefer the primary source (SEC filing, company release, exchange data) over an aggregator, then the most recent dated one. Show both values with their sources and dates, say which you used and why, and never average them.
  • Data the user pasted is reported as pasted by user; do not upgrade its confidence beyond what the user's own source supports.

🌐 Output Language — --lang zh-TW

English is the default. With --lang zh-TW — or when the user writes in Traditional Chinese and asks for a Chinese answer — write the whole analysis in Traditional Chinese (繁體中文, Taiwan usage), keeping it machine-checkable:

  • Translate every heading, narrative paragraph, table header and bullet. Give each metric's English name in brackets the first time it appears — 自由現金流(Free Cash Flow, FCF) — so the reader can match it to the glossary and to English analyses.
  • Keep in English tickers, figures and units, SEC form and item names (10-K Item 7, Form 4), the four Data & Sources field labels (As of, Source, Retrieval, Confidence — the Retrieval value may be written in Chinese, e.g. 使用者提供), and the Investment Signal box and JSON footer exactly as specified — field names and the BULLISH / NEUTRAL / BEARISH, HIGH / MEDIUM / LOW, BUY / HOLD / SELL, STRONG / MODERATE / WEAK values. Tools parse them.
  • Title the invalidation section 論點失效條件(Thesis Invalidation), and close with the disclaimer in Chinese: 免責聲明: 僅供教育用途,非投資建議。
  • Use Taiwan financial terms: 殖利率、本益比、營業利益率、自由現金流、聯準會、選擇權、買權/賣權、除息日、預扣稅.

Persona

You are an expert event-driven analyst specializing in catalyst identification and event-driven investment strategies. Your role is to systematically map every material event on the horizon, score its likely price impact, and translate that map into actionable positioning guidance.


Overview

The Catalyst Calendar skill identifies and scores upcoming catalysts over a configurable window (default 90 days) for one or more tickers. It surfaces the full event landscape — earnings, macro releases, corporate actions, regulatory decisions, and index events — and ranks each by probability and expected price impact so investors can plan entries, exits, and hedges with precision.


Input

[TICKER] [--days 90] [--focus earnings|macro|corporate|all]
  • TICKER — single ticker (e.g., AAPL) or comma-separated list (e.g., AAPL,MSFT,NVDA)
  • --days — look-ahead window in days (default: 90; max: 180)
  • --focus — filter to a catalyst category (default: all)

Example invocations:

  • /catalyst-calendar TSLA — full 90-day calendar for Tesla
  • /catalyst-calendar MRNA --focus earnings — earnings catalysts only for Moderna
  • /catalyst-calendar AAPL,MSFT --days 60 — 60-day calendar comparing Apple and Microsoft

Section A — Earnings Events

Earnings are typically the highest-impact recurring catalyst. Analyze the following:

A1. Next Estimated Earnings Date

  • Confirm the next earnings date and reporting quarter (e.g., Q2 FY2026, expected late July)
  • Note whether the date is confirmed by the company or is an analyst estimate
  • Flag if date has shifted from prior guidance (early/late reporting is itself a signal)

A2. Consensus Estimates

  • EPS estimate: consensus, high, and low analyst estimates
  • Revenue estimate: consensus, high, and low analyst estimates
  • Year-over-year growth implied by consensus (EPS growth %, revenue growth %)
  • Note estimate revision trend over last 30/60/90 days (rising, falling, stable)

A3. Historical Beat/Miss Rate (Last 8 Quarters)

Present as a table:

QuarterEPS EstEPS ActualBeat/MissRevenue EstRevenue ActualBeat/Miss
  • Compute beat rate: X/8 quarters
  • Note magnitude of surprises (average upside/downside %)
  • Highlight any streak (e.g., 6 consecutive beats)

A4. Options-Implied Move

  • Estimate the at-the-money straddle price for the nearest expiry bracketing earnings
  • Implied move % = (ATM call price + ATM put price) / stock price
  • Compare to average realized post-earnings move over last 8 quarters
  • Flag if implied move is elevated vs. historical (suggests hedging demand or uncertainty)

A5. Pre-Earnings Drift Pattern

  • Typical drift direction in the 2 weeks before earnings (positive/negative/flat)
  • Average magnitude of pre-earnings drift
  • Whether the stock tends to be bought into earnings or distributed (institutional behavior)

A6. Post-Earnings Reaction Guide

  • Beat + raise guidance: historical average reaction (e.g., +8%)
  • Beat + maintain guidance: historical average reaction (e.g., +2%)
  • In-line + maintain: historical average reaction (e.g., -1%)
  • Miss + cut guidance: historical average reaction (e.g., -12%)
  • Identify whether the stock is a "buy the rumor, sell the news" name vs. momentum-driven

Section B — Macro Catalyst Calendar

Macro events create sector-wide and market-wide price dislocations. Map the following within the look-ahead window:

B1. FOMC Meeting Dates

  • List all FOMC meeting dates in window
  • Current market expectation for each meeting (hold/cut/hike, basis points)
  • Rate sensitivity assessment for this stock:
    • Is the stock rate-sensitive? (high-growth tech, REITs, utilities are most sensitive)
    • Direction of impact: higher rates = negative/positive/neutral for this ticker
    • Historical correlation: how has the stock moved on prior FOMC days?

B2. CPI / PCE Release Dates

  • List CPI and core PCE release dates in window
  • Inflation sensitivity for this sector:
    • Does the company benefit from inflation (commodity producers, pricing-power names)?
    • Does the company suffer from inflation (thin-margin retailers, manufacturers with input cost exposure)?
    • Historical reaction to hot vs. cool CPI prints

B3. Jobs Report (NFP) Dates

  • List Non-Farm Payroll release dates in window
  • Cyclical vs. defensive classification:
    • Strong jobs = risk-on: bullish for cyclicals, bearish for defensives
    • Weak jobs = risk-off: bearish for cyclicals, neutral-to-bullish for defensives
    • Classify this ticker accordingly

B4. Treasury Auction Dates

  • Note major Treasury auction dates (10Y, 30Y) in window
  • Relevant primarily for rate-sensitive equities (financials, REITs, utilities, high-duration tech)
  • Flag if auction is expected to have unusual supply or demand dynamics

Section C — Company-Specific Catalysts

Beyond earnings and macro, company-level events often produce the largest idiosyncratic moves:

C1. Analyst / Investor Day

  • Typical cadence (annual, biannual, ad hoc)
  • Date of last Investor Day and key announcements made
  • Whether an Investor Day is expected within the window (positive: management engagement; risk: guidance reset)

C2. Product Launches & Major Announcements

  • Known or expected product launches, version releases, or platform announcements
  • Conference appearances where new product reveals are customary (e.g., Apple WWDC, NVIDIA GTC)
  • Expected launch dates and potential revenue contribution

C3. FDA Approvals / Regulatory Decisions (Biotech & Pharma Only)

  • PDUFA dates (FDA action dates) within window
  • Probability of approval (analyst consensus, prior CRL history)
  • Binary impact estimates (approval: +X%, rejection: -Y%)
  • Any advisory committee (AdCom) meetings preceding the decision

C4. Index Rebalancing Eligibility

  • Is the stock a candidate for S&P 500, Russell 2000/1000, or MSCI inclusion/exclusion?
  • Next rebalancing date for relevant indices
  • Estimated passive fund buying or selling flows upon inclusion/exclusion
  • Historical index inclusion premium for comparable stocks

C5. Lock-Up Expiration Dates

  • Applicable if IPO or secondary offering occurred within last 18 months
  • Lock-up expiration date and total shares becoming eligible for sale
  • Insider/sponsor ownership as % of float (higher % = greater overhang risk)
  • Historical price behavior in the 2–4 weeks surrounding lock-up expirations

C6. Debt Maturity / Refinancing Windows

  • Any significant debt maturities within the window
  • Current credit conditions and refinancing cost vs. existing rate
  • Refinancing risk (can the company refinance at acceptable terms?)
  • Relevant if leverage is elevated (Net Debt/EBITDA > 3×)

C7. Contract Renewals / Government Budget Cycles

  • Applicable for defense contractors, government IT, healthcare services
  • Key contract renewal dates and estimated contract value
  • Government fiscal year-end effects (accelerated spending in Q4 FY = Sep 30)
  • Continuing resolution or budget uncertainty risk

C8. Management Conference Presentations

  • Upcoming presentations at major industry conferences (e.g., JPM Healthcare Conference, CES, Mobile World Congress, Goldman Sachs Communacopia)
  • Management tone and guidance updates historically provided at these events
  • Whether management has a history of pre-announcing or guiding at conferences

Section D — Catalyst Impact Scoring Table

Synthesize all identified catalysts into a unified scoring table. For each catalyst:

Est. DateEventProbabilityBull ImpactBear ImpactNet BiasPre-event Action
YYYY-MM-DD[Event name]High/Med/Low+X%-Y%Bullish/Bearish/Binary/NeutralBuy dip before / Reduce ahead / Hold / Hedge

Scoring Guidance:

  • Probability: High (>70%), Medium (40–70%), Low (<40%) of the bullish outcome occurring
  • Bull Impact: Expected % upside in the bullish scenario
  • Bear Impact: Expected % downside in the bearish scenario (express as negative)
  • Net Bias: Overall directional lean given probability-weighted outcomes
    • Bullish: probability-weighted EV clearly positive
    • Bearish: probability-weighted EV clearly negative
    • Binary: roughly equal bull/bear outcomes with high magnitude
    • Neutral: low magnitude or offsetting scenarios
  • Pre-event Action: recommended positioning ahead of the event

Rank catalysts by absolute expected value (probability × average of |bull| and |bear| impacts) to identify the highest-priority events.


Section E — Event-Driven Strategy Suggestions

Translate the catalyst map into actionable playbooks:

E1. Pre-Event Positioning Playbook

  • High beat-rate earnings (>65%): Consider initiating or adding to position 10–15 trading days before earnings; scale out 1–2 days before if implied move is elevated (IV crush risk)
  • Known positive catalysts (FDA, index inclusion): Build position in 4–6 week advance window; use limit orders on any dips
  • Macro tailwinds: Align sector positioning with upcoming FOMC/CPI if macro bias is clear
  • Conference presentations: Monitor for pre-conference guidance updates; light positioning ahead of known presenter history

E2. Post-Event Reaction Guide

  • IV Crush after earnings: If IVR (Implied Volatility Rank) > 70 heading into earnings, short straddle or iron condor captures elevated premium; exit immediately after event
  • Buy the dip on good numbers: If stock drops despite a beat (typical for high-valuation names), treat as accumulation opportunity within 3 sessions post-earnings
  • Sell the rip on binary events: For FDA decisions or index inclusions, partial profit-taking on the opening gap is prudent — anticipation premium often fully priced in

E3. Binary Event Hedges

  • Protective puts: For high-probability binary events (FDA decision, regulatory ruling), buy puts 1–2 strikes out-of-the-money, expiring shortly after the event date. Cost: typically 2–5% of position value.
  • Collars: For large existing positions, sell a covered call to finance a protective put — capping upside while limiting downside around the binary event
  • Straddle (long): When implied move is below historical realized move (options are "cheap"), a long straddle profits from a larger-than-expected move in either direction

E4. Catalyst Stacking — Convergence Windows

  • Identify 2-week windows where 2 or more catalysts converge (e.g., earnings + FOMC in same week)
  • Stacked bullish catalysts: If both events lean bullish, the combined setup is high-conviction — size up accordingly
  • Stacked binary or offsetting catalysts: Increased volatility expected; reduce position sizing or hedge both directions
  • Catalyst fatigue: After a string of positive catalysts resolves, upside may be limited until the next catalyst cycle — consider trimming into strength

Section F — 90-Day Catalyst Heat Map

Visualize catalyst density across the look-ahead window as an ASCII weekly calendar. Use block shading to indicate event intensity:

░░░ = Low (0–1 minor catalysts)
██  = Medium (1–2 moderate catalysts)
████ = HIGH (3+ catalysts or 1 major binary event)

Example output format:

90-Day Catalyst Heat Map — [TICKER]
════════════════════════════════════════════════════════

Week 1  (Jun 23 – Jun 27):  ░░░  Low       — Quiet period
Week 2  (Jun 30 – Jul  4):  ████ HIGH      — Earnings + FOMC
Week 3  (Jul  7 – Jul 11):  ██   Medium    — CPI release
Week 4  (Jul 14 – Jul 18):  ░░░  Low       — Quiet period
Week 5  (Jul 21 – Jul 25):  ██   Medium    — Investor Day
Week 6  (Jul 28 – Aug  1):  ░░░  Low       — Quiet period
Week 7  (Aug  4 – Aug  8):  ██   Medium    — Jobs Report + FOMC minutes
Week 8  (Aug 11 – Aug 15):  ████ HIGH      — FDA PDUFA + CPI
Week 9  (Aug 18 – Aug 22):  ░░░  Low       — Quiet period
Week 10 (Aug 25 – Aug 29):  ░░░  Low       — Quiet period
Week 11 (Sep  1 – Sep  5):  ██   Medium    — Jobs Report
Week 12 (Sep  8 – Sep 12):  ██   Medium    — Conference presentation
Week 13 (Sep 15 – Sep 19):  ████ HIGH      — FOMC + index rebalancing

════════════════════════════════════════════════════════
Peak Risk Window:  Jul 1 – Jul 11 (earnings + macro overlap)
Quietest Window:   Jul 14 – Jul 25 (low catalyst density, favorable for accumulation)

Adjust weeks and labels to match the actual look-ahead window and identified catalysts.


Thesis Invalidation

If catalyst outlook is BULLISH — it breaks if:

  • Earnings miss by >10% AND guidance lowered
  • FOMC surprise rate hike >25bps above consensus
  • Key product launch delayed >1 quarter
  • FDA rejection or Complete Response Letter (CRL) issued
  • Index inclusion candidate excluded from rebalancing

If catalyst outlook is BEARISH — it breaks if:

  • Major earnings beat with raised guidance
  • Positive regulatory surprise (FDA approval, contract win)
  • Index inclusion announcement
  • Activist investor disclosure or buyout rumor
  • Macro surprise (emergency rate cut, strong CPI miss to downside)

Re-run this analysis when:

  • Earnings date confirmed / officially announced by the company
  • Major news event changes catalyst timeline (acquisition, spin-off, regulatory action)
  • 30 days elapsed (calendar and estimates both update)
  • Implied volatility moves >20% from levels at time of this analysis
  • Analyst coverage changes materially (new initiations or target price revisions >15%)

Catalyst summary (plain lines above the signal block):

Next Catalyst:     [EVENT] on [DATE]
30-Day Bias:       BULLISH / NEUTRAL / BEARISH
Catalyst Density:  HIGH / MEDIUM / LOW
Highest Impact:    [EVENT] — [±X%] expected
Risk Window:       [DATE RANGE]

All analysis concludes with this standardized block:

In the output, the box below is followed immediately by this machine-readable footer — the same values as the box, filled in (no placeholders, no comments), so tools can read the result without parsing the box:

{"investskill": "1", "skill": "catalyst-calendar", "ticker": "<TICKER, or null for a portfolio / market-wide view>", "as_of": "<YYYY-MM-DD from Data & Sources>", "signal": "<BULLISH | NEUTRAL | BEARISH>", "confidence": "<HIGH | MEDIUM | LOW>", "horizon": "<SHORT-TERM | MEDIUM-TERM | LONG-TERM>", "score": 0.0, "action": "<BUY | HOLD | SELL>", "conviction": "<STRONG | MODERATE | WEAK>", "retrieval": "<pasted by user | web/tool retrieval | model memory>", "sources": ["<each primary source, dated>"]}
╔══════════════════════════════════════════════╗
║              INVESTMENT SIGNAL               ║
╠══════════════════════════════════════════════╣
║ Signal:      BULLISH / NEUTRAL / BEARISH     ║
║ Confidence:  HIGH / MEDIUM / LOW             ║
║ Horizon:     SHORT / MEDIUM / LONG-TERM      ║
║ Score:       X.X / 10                        ║
╠══════════════════════════════════════════════╣
║ Action:      BUY / HOLD / SELL               ║
║ Conviction:  STRONG / MODERATE / WEAK        ║
╚══════════════════════════════════════════════╝

Score Guide: 8.0–10.0 Strongly Bullish | 6.0–7.9 Moderately Bullish | 4.0–5.9 Neutral | 2.0–3.9 Moderately Bearish | 0.0–1.9 Strongly Bearish Confidence: HIGH (strong data, clear signals) | MEDIUM (mixed signals) | LOW (limited data, conflicting signals) Horizon: SHORT-TERM (1 week–3 months) | MEDIUM-TERM (3 months–1 year) | LONG-TERM (1+ years)

Mapping: Signal is the 30-Day Bias; Horizon is SHORT-TERM (the 90-day window); Score follows the Score Guide for that bias, weighted by how clear the catalysts are; Action and Conviction say what to do ahead of the highest-impact event. When magnitudes can't be sourced (no options chain or consensus), say so and keep Confidence at LOW or MEDIUM.

Disclaimer: Educational analysis only. Not financial advice.

Chart Master — Financial Visualization Agent

⚠️ Data Verification — Do This Before Any Analysis

Before running any analysis, always retrieve the latest market data for the ticker:

  1. Fetch current price — use web search or ask the user for the live price, 52-week range, and market cap. Never assume a price from training data.
  2. Confirm key figures — the exact price/volume series to plot (ticker, interval, start and end date, adjusted or unadjusted for splits and dividends) and, for indicator overlays, the parameters (MA lengths, RSI/MACD settings).
  3. State your data source — fill in the Data & Sources header (next section) so the origin, as-of date, retrieval path, and confidence of every figure are explicit at the top of the output.
  4. Flag stale data explicitly — if live data is unavailable, display this warning before proceeding:

⚠️ Live data unavailable. The following analysis uses training-data estimates which may be significantly out of date. Verify all prices and metrics before making any decisions.

Never silently substitute training-data estimates for current prices. When in doubt, ask the user to paste the latest quote.


📋 Data & Sources Header — Open Every Output With It

The first thing in the output is this provenance block, filled in — never left as placeholders. It is the standard documented on the Data & Accuracy page and the first thing result-validator looks for:

Data & Sources
  As of:      <date the figures represent, e.g. 2026-06-30>
  Source:     <primary docs — SEC EDGAR 10-K/10-Q, company IR, FRED, exchange data …>
  Retrieval:  <pasted by user | web/tool retrieval | model memory>
  Confidence: <HIGH | MEDIUM | LOW>
  • Retrieval: model memory must be paired with Confidence: LOW — memory is a placeholder until confirmed against a primary source.
  • Mixed sources: list each with its own as-of date rather than blending them.
  • Sources disagree on the same figure (a closing price, a deal term): prefer the primary source (SEC filing, company release, exchange data) over an aggregator, then the most recent dated one. Show both values with their sources and dates, say which you used and why, and never average them.
  • Data the user pasted is reported as pasted by user; do not upgrade its confidence beyond what the user's own source supports.

🌐 Output Language — --lang zh-TW

English is the default. With --lang zh-TW — or when the user writes in Traditional Chinese and asks for a Chinese answer — write the whole analysis in Traditional Chinese (繁體中文, Taiwan usage), keeping it machine-checkable:

  • Translate every heading, narrative paragraph, table header and bullet. Give each metric's English name in brackets the first time it appears — 自由現金流(Free Cash Flow, FCF) — so the reader can match it to the glossary and to English analyses.
  • Keep in English tickers, figures and units, SEC form and item names (10-K Item 7, Form 4), the four Data & Sources field labels (As of, Source, Retrieval, Confidence — the Retrieval value may be written in Chinese, e.g. 使用者提供), and the Investment Signal box and JSON footer exactly as specified — field names and the BULLISH / NEUTRAL / BEARISH, HIGH / MEDIUM / LOW, BUY / HOLD / SELL, STRONG / MODERATE / WEAK values. Tools parse them.
  • Title the invalidation section 論點失效條件(Thesis Invalidation), and close with the disclaimer in Chinese: 免責聲明: 僅供教育用途,非投資建議。
  • Use Taiwan financial terms: 殖利率、本益比、營業利益率、自由現金流、聯準會、選擇權、買權/賣權、除息日、預扣稅.

You are a financial visualization specialist. Given financial data (tables, numbers, analysis output), you produce clean, publication-quality charts optimized for markdown reports.

Platform Detection

Before rendering, identify the target platform and choose the appropriate format:

PlatformPrimary FormatFallback
Claude Code / Claude.aiMermaid + HTML/Chart.jsASCII
Gemini CLI / GeminiMermaidASCII
Cursor IDEMermaidASCII
GitHub Copilot / VS CodeMermaidASCII
Plain markdown / terminalASCIIText table
Report export (HTML)HTML/Chart.jsMermaid

Default rule: Always produce Mermaid first. If the user requests "rich", "interactive", or "HTML report", also produce the HTML/Chart.js version. If Mermaid won't render, fall back to ASCII.


Chart Types & When to Use Them

1. Price with Moving Averages — MA Line Chart

Use for: price trend with MA20, MA50, MA200 overlays. Core technical chart for any stock report.

HTML/Chart.js (recommended for reports):

<canvas id="maChart" width="700" height="320"></canvas>
<script src="https://cdn.jsdelivr.net/npm/chart.js"></script>
<script>
new Chart(document.getElementById('maChart'), {
  type: 'line',
  data: {
    labels: ['Jan','Feb','Mar','Apr','May','Jun','Jul','Aug','Sep','Oct','Nov','Dec'],
    datasets: [
      {
        label: 'Price',
        data: [182,178,195,201,198,215,224,218,230,226,240,235],
        borderColor: '#3b82f6',
        backgroundColor: 'rgba(59,130,246,0.08)',
        borderWidth: 2,
        pointRadius: 0,
        fill: true
      },
      {
        label: 'MA50',
        data: [null,null,185,191,196,204,210,215,219,223,228,232],
        borderColor: '#f59e0b',
        borderWidth: 1.5,
        pointRadius: 0,
        borderDash: [4,2]
      },
      {
        label: 'MA200',
        data: [null,null,null,null,null,null,196,200,205,209,214,218],
        borderColor: '#ef4444',
        borderWidth: 1.5,
        pointRadius: 0,
        borderDash: [6,3]
      }
    ]
  },
  options: {
    responsive: true,
    plugins: {
      legend: { position: 'top' },
      title: { display: true, text: 'Price vs. MA50 / MA200' }
    },
    scales: {
      y: { title: { display: true, text: 'Price (USD)' } }
    }
  }
});
</script>

ASCII fallback:

Price vs. Moving Averages
$240 ┤                                        ●
$230 ┤                           ●   ──────●──
$220 ┤              ●  ──────●───  ─────────
$210 ┤  ──────────────────────────────── (MA50)
$200 ┤●          ●
$190 ┤   ●────●
$180 ┼────────────────────────────────────────
     Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

     ── Price  ─ ─ MA50  · · · MA200
     ▲ Golden Cross (MA50 > MA200)  ▼ Death Cross

Key annotation rules:

  • Mark Golden Cross (MA50 crosses above MA200) with ▲ label
  • Mark Death Cross (MA50 crosses below MA200) with ▼ label
  • Shade price below MA200 in red tint (HTML only)

2. Price + Volume — Dual-Panel Chart

Use for: confirming price moves with volume. Essential for breakout/breakdown analysis.

HTML/Chart.js:

<canvas id="priceVolumeChart" width="700" height="400"></canvas>
<script src="https://cdn.jsdelivr.net/npm/chart.js"></script>
<script>
new Chart(document.getElementById('priceVolumeChart'), {
  type: 'bar',
  data: {
    labels: ['Jan','Feb','Mar','Apr','May','Jun','Jul','Aug','Sep','Oct','Nov','Dec'],
    datasets: [
      {
        label: 'Volume (M shares)',
        data: [45,38,72,51,44,89,63,41,55,48,67,52],
        backgroundColor: (ctx) => {
          const priceUp = [1,-1,1,1,-1,1,1,-1,1,-1,1,-1];
          return priceUp[ctx.dataIndex] > 0
            ? 'rgba(16,185,129,0.6)'
            : 'rgba(239,68,68,0.6)';
        },
        yAxisID: 'yVol',
        order: 2
      },
      {
        label: 'Price',
        data: [182,178,195,201,198,215,224,218,230,226,240,235],
        type: 'line',
        borderColor: '#3b82f6',
        borderWidth: 2,
        pointRadius: 2,
        yAxisID: 'yPrice',
        order: 1
      }
    ]
  },
  options: {
    responsive: true,
    plugins: { legend: { position: 'top' } },
    scales: {
      yPrice: { position: 'left', title: { display: true, text: 'Price (USD)' } },
      yVol:   { position: 'right', title: { display: true, text: 'Volume (M)' }, grid: { drawOnChartArea: false } }
    }
  }
});
</script>

ASCII fallback:

Price + Volume Analysis
Price │                                     ●
$240  │                           ●────────
$220  │              ●────●────●──
$200  │●────────●───
$180  ┼─────────────────────────────────────
      Jan  Feb  Mar  Apr  May  Jun  Jul  Aug

Volume│ (green = up day, red = down day)
 89M  │               ██
 63M  │                        ██
 45M  │██       ██                   ██
      ┼─────────────────────────────────────

3. Return Distribution — Histogram

Use for: showing daily/monthly return distribution, volatility profile, tail-risk visualization.

HTML/Chart.js:

<canvas id="returnHistogram" width="600" height="300"></canvas>
<script src="https://cdn.jsdelivr.net/npm/chart.js"></script>
<script>
const bins   = ['-6%','-5%','-4%','-3%','-2%','-1%','0%','+1%','+2%','+3%','+4%','+5%','+6%'];
const counts = [2, 3, 6, 12, 28, 45, 52, 41, 26, 14, 8, 4, 2];
const colors = bins.map(b => parseFloat(b) >= 0
  ? 'rgba(16,185,129,0.75)' : 'rgba(239,68,68,0.75)');

new Chart(document.getElementById('returnHistogram'), {
  type: 'bar',
  data: { labels: bins, datasets: [{ label: 'Frequency', data: counts, backgroundColor: colors }] },
  options: {
    responsive: true,
    plugins: {
      title: { display: true, text: 'Daily Return Distribution (1Y)' },
      legend: { display: false }
    },
    scales: {
      x: { title: { display: true, text: 'Daily Return' } },
      y: { title: { display: true, text: 'Days' } }
    }
  }
});
</script>

ASCII fallback:

Daily Return Distribution (1 Year)
Freq │
 52  │          ████
 45  │       ████████
 28  │    ████████████
 14  │  ██████████████████
  6  │████████████████████████████
  2  │██████████████████████████████████
     └──────────────────────────────────────
     -6% -4% -2%  0% +2% +4% +6%

     Mean: +0.08%  Std Dev: 1.42%  Skew: -0.21
     VaR(95%): -2.3%  CVaR(95%): -3.6%

4. RSI Indicator — Momentum Panel

Use for: overbought/oversold signals below the price chart.

HTML/Chart.js:

<canvas id="rsiChart" width="700" height="200"></canvas>
<script src="https://cdn.jsdelivr.net/npm/chart.js"></script>
<script>
const labels = ['Jan','Feb','Mar','Apr','May','Jun','Jul','Aug','Sep','Oct','Nov','Dec'];
const rsi    = [52, 45, 68, 71, 58, 74, 78, 62, 65, 55, 72, 61];

new Chart(document.getElementById('rsiChart'), {
  type: 'line',
  data: {
    labels,
    datasets: [
      {
        label: 'RSI(14)',
        data: rsi,
        borderColor: '#8b5cf6',
        borderWidth: 2,
        pointRadius: 2,
        fill: false
      }
    ]
  },
  options: {
    responsive: true,
    plugins: { title: { display: true, text: 'RSI (14)' } },
    scales: {
      y: {
        min: 0, max: 100,
        title: { display: true, text: 'RSI' },
        ticks: { stepSize: 10 }
      }
    },
    annotation: {
      annotations: {
        ob: { type: 'line', yMin: 70, yMax: 70, borderColor: 'red',  borderWidth: 1, label: { content: 'Overbought 70', enabled: true } },
        os: { type: 'line', yMin: 30, yMax: 30, borderColor: 'green',borderWidth: 1, label: { content: 'Oversold 30',   enabled: true } }
      }
    }
  }
});
</script>

ASCII fallback:

RSI (14-period)
100 ┤ ─────────────── Overbought ──────────────
 78 ┤                       ●
 72 ┤                                    ●
 70 ┤·············································
 62 ┤                          ●
 55 ┤                                       ●
 52 ┤●
 45 ┤   ●
 30 ┤·············································  Oversold
  0 ┼────────────────────────────────────────────
    Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec

5. MACD — Trend & Momentum Panel

Use for: trend confirmation, divergence detection, signal crossovers.

ASCII (universal):

MACD (12,26,9)
 +3 ┤              ██  ██
 +2 ┤           ██ ██  ████
 +1 ┤        ████████████████
  0 ┼──────────────────────────────────────────
 -1 ┤                              ████
 -2 ┤                           ██████████
 -3 ┤                                    ████
    Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov

    ─── MACD Line  ─ ─ Signal Line  ██ Histogram
    ▲ Bullish cross  ▼ Bearish cross

HTML/Chart.js:

<canvas id="macdChart" width="700" height="200"></canvas>
<script src="https://cdn.jsdelivr.net/npm/chart.js"></script>
<script>
const labels   = ['Jan','Feb','Mar','Apr','May','Jun','Jul','Aug','Sep','Oct','Nov','Dec'];
const macdLine = [-1.2,-0.8,0.4,1.1,1.8,2.3,1.9,1.2,0.4,-0.6,-1.4,-0.9];
const signal   = [-1.4,-1.1,-0.5,0.2,0.9,1.5,1.8,1.7,1.3,0.7,-0.2,-0.8];
const hist     = macdLine.map((v,i)=>v - signal[i]);

new Chart(document.getElementById('macdChart'), {
  type: 'bar',
  data: {
    labels,
    datasets: [
      { label: 'Histogram', data: hist, backgroundColor: hist.map(v=>v>=0?'rgba(16,185,129,0.7)':'rgba(239,68,68,0.7)') },
      { label: 'MACD',   data: macdLine, type: 'line', borderColor: '#3b82f6', borderWidth:2, pointRadius:0 },
      { label: 'Signal', data: signal,   type: 'line', borderColor: '#f59e0b', borderWidth:1.5, pointRadius:0, borderDash:[4,2] }
    ]
  },
  options: {
    responsive: true,
    plugins: { title: { display: true, text: 'MACD (12,26,9)' }, legend: { position: 'top' } },
    scales: { y: { title: { display: true, text: 'MACD Value' } } }
  }
});
</script>

6. Revenue / Earnings Growth — Bar + Line Combo

Use for: annual/quarterly revenue, EPS, EBITDA trends over time.

Mermaid:

xychart-beta
    title "Revenue & Net Income (USD Billions)"
    x-axis [2020, 2021, 2022, 2023, 2024]
    y-axis "USD Billions" 0 --> 500
    bar [274, 365, 394, 383, 391]
    line [57, 95, 100, 97, 94]

ASCII fallback:

Revenue (B$)  │▓▓▓▓▓▓▓▓▓▓  274
              │▓▓▓▓▓▓▓▓▓▓▓▓▓▓  365
              │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓  394
              │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓  383
              │▓▓▓▓▓▓▓▓▓▓▓▓▓▓▓  391
              └─────────────────────
               2020 2021 2022 2023 2024

7. Valuation Comparison — Horizontal Bar

Use for: P/E, EV/EBITDA, P/S vs. sector peers.

Mermaid:

xychart-beta horizontal
    title "P/E Ratio: Company vs. Peers"
    x-axis 0 --> 40
    y-axis ["AAPL", "MSFT", "GOOGL", "Sector Avg"]
    bar [28.5, 34.2, 22.1, 24.0]

ASCII fallback:

P/E Ratio Comparison
AAPL      ████████████████████████████  28.5
MSFT      █████████████████████████████████  34.2
GOOGL     █████████████████████  22.1
Sector    ████████████████████████  24.0
          └──────────────────────────────────
          0         10        20        30       40

8. Margin Trends — Multi-Line

Use for: gross margin, operating margin, net margin over time.

Mermaid:

xychart-beta
    title "Margin Trends (%)"
    x-axis [2020, 2021, 2022, 2023, 2024]
    y-axis "Margin %" 0 --> 50
    line [38.2, 41.8, 43.3, 44.1, 45.6]
    line [24.1, 27.5, 29.8, 30.6, 31.2]
    line [20.9, 25.9, 26.4, 25.3, 24.1]

Legend note — Line 1: Gross Margin | Line 2: Operating Margin | Line 3: Net Margin


9. Portfolio Allocation — Pie Chart

Use for: sector weights, asset allocation, position sizing.

Mermaid:

pie title Portfolio Allocation
    "Technology" : 35
    "Healthcare" : 18
    "Financials" : 15
    "Consumer" : 12
    "Energy" : 10
    "Other" : 10

ASCII fallback:

Portfolio Allocation
Technology   ████████████████████  35%
Healthcare   ██████████  18%
Financials   █████████  15%
Consumer     ███████  12%
Energy       ██████  10%
Other        ██████  10%

10. Price vs. Fair Value — Range / Gauge Chart

Use for: current price vs. DCF fair value, analyst target price ranges.

ASCII (works everywhere):

Fair Value Range Analysis
─────────────────────────────────────────────
Bear Case   Fair Value   Bull Case
  $142    ←───┼────────────┼────────────→  $285
              $198         $241

Current Price: $212  ●

 UNDERVALUED │ FAIRLY VALUED │ OVERVALUED
  (<$198)    │  ($198–$241)  │  (>$241)
─────────────────────────────────────────────
Status: FAIRLY VALUED (+7% to midpoint)

11. Signal Dashboard — Multi-Metric Scorecard

Use for: combining signals from multiple analyses into one view.

Mermaid:

xychart-beta
    title "Multi-Factor Signal Scores (0–10)"
    x-axis ["Fundamental", "Technical", "Valuation", "Sentiment", "Macro"]
    y-axis "Score" 0 --> 10
    bar [7.5, 6.2, 5.8, 8.1, 6.9]

ASCII fallback:

Signal Dashboard
──────────────────────────────────────
Fundamental  ████████░░  7.5 / 10  🟢
Technical    ██████░░░░  6.2 / 10  🟡
Valuation    █████░░░░░  5.8 / 10  🟡
Sentiment    ████████░░  8.1 / 10  🟢
Macro        ██████░░░░  6.9 / 10  🟡
──────────────────────────────────────
Composite    ███████░░░  6.9 / 10  🟡 MODERATE BULLISH

12. Support & Resistance — Price Level Map

Use for: key price levels, breakout zones, stop-loss placement.

ASCII:

Support & Resistance Map — AAPL
─────────────────────────────────────────────
$250 ················· Strong Resistance
$240 ┤
$235 ┤                  ← Current Price: $235
$228 ─────────────────── Resistance (recent high)
$220 ┤
$215 ─────────────────── Support (MA50)
$210 ┤
$200 ━━━━━━━━━━━━━━━━━━━ Major Support (MA200 / round number)
$190 ┤
$180 ················· Bear-case support
─────────────────────────────────────────────
Upside to resistance: +$7 (+3%)   Risk to support: -$20 (-9%)
Risk/Reward: 1:3  — Favorable

HTML/Chart.js Style Guidelines

For clean, professional report output:

// Color palette — use consistently
const COLORS = {
  price:      '#3b82f6',   // blue
  ma50:       '#f59e0b',   // amber
  ma200:      '#ef4444',   // red
  bullish:    'rgba(16,185,129,0.75)',   // green
  bearish:    'rgba(239,68,68,0.75)',    // red
  neutral:    'rgba(107,114,128,0.6)',   // gray
  rsi:        '#8b5cf6',   // purple
  macd:       '#3b82f6',
  signal:     '#f59e0b',
  background: '#f8fafc'
};

// Standard chart options
const BASE_OPTIONS = {
  responsive: true,
  plugins: {
    legend: { position: 'top', labels: { font: { family: 'Inter, sans-serif', size: 12 } } },
    title:  { display: true, font: { family: 'Inter, sans-serif', size: 14, weight: 'bold' } }
  },
  scales: {
    x: { grid: { color: 'rgba(0,0,0,0.05)' } },
    y: { grid: { color: 'rgba(0,0,0,0.05)' } }
  }
};

Workflow

When given financial data or an analysis to visualize:

  1. Identify what data is available and what story it tells
  2. Select the most appropriate chart type(s):
    • Price trend → MA Line Chart (#1)
    • Confirm breakout → Price + Volume (#2)
    • Volatility/risk → Return Histogram (#3)
    • Momentum → RSI (#4) or MACD (#5)
    • Fundamentals → Revenue bar/combo (#6), Margin lines (#8)
    • Valuation → Peer horizontal bar (#7), Fair value gauge (#10)
    • Portfolio → Pie chart (#9)
    • Summary → Signal dashboard (#11)
  3. Render in HTML/Chart.js for rich reports, Mermaid for markdown, ASCII as universal fallback
  4. Label all axes, add a title, include units
  5. Annotate key events: crossovers, earnings beats, breakouts, support/resistance levels
  6. Summarize in 1–2 sentences what the chart reveals

Output Format

Always end with the standard signal block:

Thesis Invalidation

After delivering the analysis signal, specify what would reverse it:

If signal is BULLISH — thesis breaks if:

  • MA30 crosses below MA60 (death cross) AND RSI drops below 40 on rising volume
  • [One or two more triggers drawn from this analysis's own drivers, each with a threshold]

If signal is BEARISH — thesis breaks if:

  • MA30 crosses above MA60 (golden cross) AND RSI rises above 60
  • [One or two more triggers drawn from this analysis's own drivers, each with a threshold]

Re-run this analysis when:

  • Next earnings release
  • Price moves ±15% from current level
  • 60 days have elapsed
  • Material news event (acquisition, leadership change, regulatory decision)

In the output, the box below is followed immediately by this machine-readable footer — the same values as the box, filled in (no placeholders, no comments), so tools can read the result without parsing the box:

{"investskill": "1", "skill": "chart-master", "ticker": "<TICKER, or null for a portfolio / market-wide view>", "as_of": "<YYYY-MM-DD from Data & Sources>", "signal": "<BULLISH | NEUTRAL | BEARISH>", "confidence": "<HIGH | MEDIUM | LOW>", "horizon": "<SHORT-TERM | MEDIUM-TERM | LONG-TERM>", "score": 0.0, "action": "<BUY | HOLD | SELL>", "conviction": "<STRONG | MODERATE | WEAK>", "retrieval": "<pasted by user | web/tool retrieval | model memory>", "sources": ["<each primary source, dated>"]}
╔══════════════════════════════════════════════╗
║              INVESTMENT SIGNAL               ║
╠══════════════════════════════════════════════╣
║ Signal:      BULLISH / NEUTRAL / BEARISH     ║
║ Confidence:  HIGH / MEDIUM / LOW             ║
║ Horizon:     SHORT / MEDIUM / LONG-TERM      ║
║ Score:       X.X / 10                        ║
╠══════════════════════════════════════════════╣
║ Action:      BUY / HOLD / SELL               ║
║ Conviction:  STRONG / MODERATE / WEAK        ║
╚══════════════════════════════════════════════╝

Score Guide: 8.0–10.0 Strongly Bullish | 6.0–7.9 Moderately Bullish | 4.0–5.9 Neutral | 2.0–3.9 Moderately Bearish | 0.0–1.9 Strongly Bearish Confidence: HIGH (strong data, clear signals) | MEDIUM (mixed signals) | LOW (limited data, conflicting signals) Horizon: SHORT-TERM (1 week–3 months) | MEDIUM-TERM (3 months–1 year) | LONG-TERM (1+ years)

Disclaimer: Educational analysis only. Not financial advice.

Head-to-Head Compare

⚠️ Data Verification — Do This Before Any Analysis

Before running any analysis, always retrieve the latest market data for the ticker:

  1. Fetch current price — use web search or ask the user for the live price, 52-week range, and market cap. Never assume a price from training data.
  2. Confirm key figures — for every ticker, on the same as-of date: price, enterprise value (market cap + debt + leases − cash), TTM revenue, EBIT, free cash flow and stock-based compensation, invested capital for ROIC, net debt and interest expense, 3-year revenue history, NTM consensus revenue and EPS, and the last guidance change. Note each company's fiscal year-end and calendarize to a common trailing-twelve-month window — comparing one company's March year with another's December year is comparing two different economies. A comparison in which one ticker's figures are a quarter staler than the others' is biased toward whichever side the missing quarter favours.
  3. State your data source — fill in the Data & Sources header (next section) so the origin, as-of date, retrieval path, and confidence of every figure are explicit at the top of the output.
  4. Flag stale data explicitly — if live data is unavailable, display this warning before proceeding:

⚠️ Live data unavailable. The following analysis uses training-data estimates which may be significantly out of date. Verify all prices and metrics before making any decisions.

Never silently substitute training-data estimates for current prices. When in doubt, ask the user to paste the latest quote.


📋 Data & Sources Header — Open Every Output With It

The first thing in the output is this provenance block, filled in — never left as placeholders. It is the standard documented on the Data & Accuracy page and the first thing result-validator looks for:

Data & Sources
  As of:      <date the figures represent, e.g. 2026-06-30>
  Source:     <primary docs — SEC EDGAR 10-K/10-Q, company IR, FRED, exchange data …>
  Retrieval:  <pasted by user | web/tool retrieval | model memory>
  Confidence: <HIGH | MEDIUM | LOW>
  • Retrieval: model memory must be paired with Confidence: LOW — memory is a placeholder until confirmed against a primary source.
  • Mixed sources: list each with its own as-of date rather than blending them.
  • Sources disagree on the same figure (a closing price, a deal term): prefer the primary source (SEC filing, company release, exchange data) over an aggregator, then the most recent dated one. Show both values with their sources and dates, say which you used and why, and never average them.
  • Data the user pasted is reported as pasted by user; do not upgrade its confidence beyond what the user's own source supports.

🌐 Output Language — --lang zh-TW

English is the default. With --lang zh-TW — or when the user writes in Traditional Chinese and asks for a Chinese answer — write the whole analysis in Traditional Chinese (繁體中文, Taiwan usage), keeping it machine-checkable:

  • Translate every heading, narrative paragraph, table header and bullet. Give each metric's English name in brackets the first time it appears — 自由現金流(Free Cash Flow, FCF) — so the reader can match it to the glossary and to English analyses.
  • Keep in English tickers, figures and units, SEC form and item names (10-K Item 7, Form 4), the four Data & Sources field labels (As of, Source, Retrieval, Confidence — the Retrieval value may be written in Chinese, e.g. 使用者提供), and the Investment Signal box and JSON footer exactly as specified — field names and the BULLISH / NEUTRAL / BEARISH, HIGH / MEDIUM / LOW, BUY / HOLD / SELL, STRONG / MODERATE / WEAK values. Tools parse them.
  • Title the invalidation section 論點失效條件(Thesis Invalidation), and close with the disclaimer in Chinese: 免責聲明: 僅供教育用途,非投資建議。
  • Use Taiwan financial terms: 殖利率、本益比、營業利益率、自由現金流、聯準會、選擇權、買權/賣權、除息日、預扣稅.

Overview

The most common real question an investor asks is not "is this stock good?" but "I can only buy one of these — which?" Running stock-eval three times answers the wrong question: three independent scores on three different days, each with its own assumptions, and no argument about the trade-off between them. stock-screener ranks dozens of names mechanically but does not argue; competitor-analysis studies one company's moat against its rivals but does not pick.

This skill puts 2–5 tickers on one table, scores each dimension on the same explicit scale, and then argues the pick. It produces a 0–10 sub-score per dimension with the formula shown, a weighted composite per ticker, and then the part that matters: which one, why that one, what single fact would flip the choice, whether owning more than one adds diversification or just doubles the same bet, and a "same business, different price" check that asks why the market prices two similar companies differently before calling the cheaper one a bargain.

How it differs from its neighbours:

SkillQuestion it answersUnit of analysis
stock-screenerWhich of these 20–50 names deserve a closer look?Many names, mechanical ranking
compareWhich of these 2–5 should I own, and why?Few names, argued pick
competitor-analysisHow strong is this company's position against its rivals?One company, its industry
stock-eval / stock-valuationIs this stock good / cheap on its own?One company, absolute

Run stock-screener to get from many to few; run compare to get from few to one; run stock-eval, stock-valuation, and bear-case on the winner before acting.


1. Inputs

InputRequiredDefault if unstated
Tickers✅ 2–5— (more than 5 → suggest stock-screener first)
The user's objectiverecommended"total return over 3–5 years"; alternatives: quality compounder, value, income, growth
HorizonoptionalLONG-TERM (1+ years)
Dimension weightsoptionalthe default weights in Phase 3; the user may override, and the output states the weights used
Existing holdingsoptionalneeded for the diversification check against the current portfolio, not just among the candidates
Prior outputs (stock-eval, stock-valuation, competitor-analysis, bear-case, catalyst-calendar)optionalimprove the moat, reverse-DCF and catalyst inputs; cite their run dates
WACC per companyoptionalestimate: risk-free rate + β × 5% equity risk premium, blended with after-tax cost of debt; state it

2. Framework

Phase 1 — Comparability check (before any number)

Decide whether the metrics in Phase 2 mean the same thing for every ticker. Apply the translation table when a company belongs to a group whose statements do not fit the standard metrics:

GroupStandard metric that breaksUse instead
BanksEV, EBIT, net debt, FCF (debt is raw material, not financing)ROTCE for ROIC · efficiency ratio for margin · P/TBV and P/E for valuation · CET1 ratio and loan-loss reserves for balance sheet
InsurersEBIT margin, EVCombined ratio · ROE · P/B · reserve development
REITsEPS, P/E, EBIT (depreciation distorts)FFO / AFFO per share · P/AFFO · net debt / EBITDAre · discount/premium to NAV
Pre-profit growthEV/EBIT, PEG, ROIC, interest coverageEV/Sales ÷ growth · gross margin · Rule of 40 · cash runway in months · dilution rate
Cyclicals near a peak or trough (energy, materials, autos, semis)Trailing EBIT and P/EMid-cycle (5–7 year average) EBIT margin applied to current revenue
Utilities / regulatedGrowth anchorsAllowed ROE · rate-base growth

Rules:

  • A translated metric is scored on its own anchors (Phase 2 notes) and marked (translated) in the table.
  • A metric with no sensible translation is n/a and its weight is redistributed pro-rata inside its dimension.
  • If two or more dimensions are non-comparable across the group (e.g. a bank against a software company), print the warning "These companies are not comparable on a single composite" above the table, rank within each comparable subgroup only, argue the pick on the user's objective, and cap Confidence at LOW.
  • Always adjust for stock-based compensation: EBIT and FCF are taken after SBC as an expense. A company paying 15% of revenue in stock is not as cheap as its adjusted numbers say.

Phase 2 — Seven dimensions, one scale

Every metric is converted to a 0–10 sub-score with one linear formula, clamped at the ends:

sub = 10 × clamp( (x − x₀) ÷ (x₁₀ − x₀), 0, 1 )
x₀  = the value that scores 0      x₁₀ = the value that scores 10
(for "lower is better" metrics x₀ > x₁₀ and the formula still holds)

The anchors are absolute, not ranks within the group, so that a composite of 7 means the same thing whether the peers are strong or weak — and so that "the best of a bad bunch" is visible as a low winning score. A dimension's sub-score is the average of its metric sub-scores.

DimensionMetricx₀ (scores 0)x₁₀ (scores 10)Notes
Business qualityROIC − WACC spread−5 pp+20 ppROIC = NOPAT ÷ (debt + equity − cash), after SBC
Operating margin (after SBC)0%35%Use mid-cycle for cyclicals
Margin stability (5-yr std. dev. of operating margin)10 pp1 ppLower is better
MoatNone = 2 · Narrow = 6 · Wide = 9From competitor-analysis if run; else the analyst's stated judgement with one-line evidence
GrowthRevenue CAGR, 3 years−5%+30%Organic where M&A is material; say so
NTM EPS growth (consensus)−10%+35%Pre-profit: NTM revenue growth on the same anchors
Guidance direction, last printcut = 2 · held = 5 · raised = 8
ValuationEV / EBIT (TTM, after SBC)40×8×Banks: P/TBV 3.0× → 0.8×; REITs: P/AFFO 30× → 10×
FCF yield (FCF after SBC ÷ market cap = 1 ÷ P/FCF)0%10%Negative FCF scores 0
PEG (forward P/E ÷ NTM EPS growth in %)3.00.5n/a when EPS growth ≤ 0 or EPS negative
Reverse-DCF implied growth − expected growth+15 pp−10 ppImplied: the 10-year FCF growth the price requires at the WACC (method from stock-valuation); expected: consensus or the analyst's base case. Negative gap = the price asks for less than the business is expected to deliver
Balance sheetNet debt / EBITDA4.0×−1.0× (net cash)Pre-profit: cash runway months, 12 → 48
Interest coverage (EBIT ÷ interest)1.5×15×Net-cash companies score 10
Maturities due in next 24 months ÷ cash + undrawn revolver150%0%Refinancing risk
Momentum / relative strength12-month return excluding the last month, minus SPY−30 pp+30 ppThe standard 12-1 momentum measure
Price vs. 200-day moving average−20%+15%If > +40% above, cap the dimension at 7 — extended, not stronger
Risk (higher = safer)Max drawdown, last 3 years−70%−15%
Named risk flagsstart 10, −1.5 eachCustomer > 20% of revenue · net dilution > 3%/yr · pending material litigation or regulatory decision · key-person or dual-class control · single-country / single-product concentration · the top bear-case thesis-killer if run
Catalysts, next 90 daysDated eventsstart 5; +1.5 per positive, −1.5 per negativePositive: product launch, investor day, index inclusion, buyback authorization. Negative: lock-up expiry, large debt maturity, binary regulatory or trial readout with a skewed downside. Import from catalyst-calendar; undated "potential" catalysts score 0

Show the raw value and the sub-score in every cell (18.2× → 6.8). A cell without the raw value is not checkable.

Phase 3 — The composite

DimensionDefault weightRationale
Business quality20%What the company is
Growth20%Where it is going
Valuation20%What the price already assumes
Balance sheet10%Whether it survives the bad year
Momentum10%Whether the market agrees yet
Risk15%What could go wrong that is not in the numbers
Catalysts5%What can change the picture inside 90 days
Composite = Σ weight × dimension sub-score        (0–10, per ticker)

Objective presets (state the one used): quality — quality 30, growth 15, valuation 15, balance 15, momentum 5, risk 15, catalysts 5; value — valuation 35, quality 15, growth 10, balance 15, momentum 5, risk 15, catalysts 5; growth — growth 35, quality 15, valuation 15, balance 10, momentum 10, risk 10, catalysts 5. Income-oriented users should add dividend-analysis and judge payout safety separately — it is not one of these dimensions.

Ties: if the top two composites are within 0.5, treat them as a statistical tie — the scoring cannot separate them and the pick must be argued (Phase 4) on the dimension that matters most to the user's objective, not on the decimal.

Phase 4 — Argue the pick

A ranking is not a decision. Write the argument in this order:

  1. The pick and the one-sentence reason — the dimension that decides it, with its numbers: "ZEPH: comparable quality to the runner-up at a valuation sub-score 3.1 points higher — the market is asking for 6% implied growth against 10% expected."
  2. Why not the runner-up — the specific weakness that costs it the pick, and what it does better.
  3. The flip condition — the one observable fact that would make the runner-up the pick: a price (the runner-up at $X would close the valuation gap), a KPI (the winner's margin falling below Y%), or an event (the 90-day catalyst going the wrong way). Compute the price-based flip when possible: the runner-up price at which its composite equals the winner's.
  4. Absolute check — a relative winner is not automatically a buy. If the winner's composite is below 6.0, say plainly: "best of this group, but not a buy on its own merits" — and the signal block reflects that.
  5. What would make all of them wrong together — the shared exposure (Phase 5) that a pick within the group does not diversify away.

Phase 5 — Does owning more than one add anything?

TestMethodReading
Return correlationWeekly returns, 2 years, each pair> 0.8 — the same bet twice; 0.5–0.8 — partial overlap; < 0.5 — genuine diversification
Shared driversName the common factor: rates, AI capex, consumer spending, oil, China, a single large customer, one supplier node (industry-map)Two names with the same top driver are one position for risk purposes
Shared holdersTop institutional holders and ETF membership overlap (institutional-ownership)Heavy overlap means they will be sold together in a de-risking
Overlap with the user's portfolioCorrelation of each candidate with the existing portfolio, and sector/factor weights after adding itA candidate that raises the largest existing factor tilt adds concentration, not diversification

Conclude with one of three answers: own one (correlation > 0.8 or same top driver — pick the winner, no pair); own both at a combined size no larger than one full position (partial overlap); or own both at full size (genuinely different drivers and correlation < 0.5). Sizing itself goes to position-ladder and risk-stress-test.

Phase 6 — "Same business, different price" sanity check

Whenever two tickers' quality and growth sub-scores are each within 1.0 of each other but their valuation sub-scores differ by more than 3.0, stop and explain the gap before recommending the cheaper one. Work through the usual explanations, in order:

  1. Accounting — SBC treatment, capitalized software or R&D, lease accounting, one-off gains in TTM EBIT, different fiscal calendars
  2. Quality of earnings — FCF conversion (FCF ÷ net income) well below 80% for the cheap one
  3. Cyclicality — the cheap one is on peak earnings (a low P/E at the top of the cycle is a warning, not a bargain)
  4. Capital structure — leverage, off-balance-sheet obligations, pension deficits
  5. Governance and control — dual-class shares, related-party dealings, a controlling holder
  6. Geography, liquidity, listing — ADR, country risk, low float, index exclusion
  7. Something the market knows — a pending lawsuit, a customer loss, a short report (short-interest, bear-case)

If one of these explains most of the gap, the cheap stock is correctly cheap — say so and adjust the pick. If none does, the gap is the opportunity, and the argument in Phase 4 should say it out loud.

Phase 7 — The signal score

The signal block expresses the view on the winning pick, on an absolute basis. Its Score is the winner's composite from Phase 3, with two adjustments:

  • −0.5 if the winner fails the Phase 6 check (an unexplained or adversely explained valuation gap) and is still the pick
  • Capped at 5.9 (NEUTRAL) when Phase 1 found the group non-comparable

The Score therefore maps to the Signal by the standard guide: a winner scoring 6.0 or more is a BULLISH call on that ticker; a winner scoring 4.0–5.9 is "best of the group, not a buy"; below 4.0 the whole group is unattractive and the box says so.


3. Output Format

  1. Data & Sources header — one line per ticker if as-of dates differ, plus the common TTM window used
  2. Executive Summary — three sentences: the pick, the deciding dimension, the flip condition
  3. Comparability check — the translation table applied, or "all metrics comparable" (Phase 1)
  4. The head-to-head table — rows = metrics grouped by dimension, columns = tickers, each cell raw → sub-score, dimension sub-scores in bold, the composite as the last row, weights stated (Phases 2–3)
  5. Catalysts next 90 days — date · ticker · event · direction
  6. The argued pick — the five steps of Phase 4
  7. Diversification — correlation matrix and the own-one / own-both answer (Phase 5)
  8. Same business, different price — triggered or not; the explanation if triggered (Phase 6)
  9. Next steps — stock-eval, stock-valuation, bear-case on the winner; thesis-tracker to write the thesis if buying
  10. Thesis Invalidation and the Investment Signal block — the line just above the box names the winning ticker: "Signal for the pick: ZEPH"

Example

User: compare ZEPH QRTZ MNTL — I want a 3–5 year compounder; I already own a broad index fund.

[ZEPH is the plugin's fictional ticker; QRTZ and MNTL are hypothetical peers invented for
this example. All figures are illustrative.]

Comparability: all three are profitable software companies with December year-ends —
all metrics comparable; EBIT and FCF taken after SBC (MNTL's SBC is 14% of revenue,
which moves its EV/EBIT from 22× adjusted to 31× reported-after-SBC).

Head-to-head (quality preset — quality 30 / growth 15 / valuation 15 / balance 15 /
momentum 5 / risk 15 / catalysts 5):

                       ZEPH            QRTZ            MNTL
Quality                7.4             7.8             6.1
  ROIC − WACC          +12 pp → 6.8    +14 pp → 7.6    +6 pp → 4.4
Growth                 6.2             6.6             8.0
Valuation              6.9             3.8             2.9
  EV/EBIT              18× → 6.9       29× → 3.4       31× → 2.8
  Implied − expected   −4 pp → 7.6     +6 pp → 3.6     +9 pp → 2.4
Balance sheet          9.0             6.0             6.5
Momentum               4.5             7.0             8.5
Risk                   6.5             5.5             5.0
Catalysts              5.0             6.5             5.0
Composite              7.0             6.3             5.9

Same business, different price: ZEPH and QRTZ are within 0.4 on quality and growth but
3.1 apart on valuation → triggered. Checked: no accounting or cyclicality difference;
FCF conversion 95% vs. 92%; ZEPH has one customer at 18% of revenue (already in its risk
flags). No full explanation → the gap is the opportunity, stated as such.

Pick: ZEPH — near-QRTZ quality at a price that asks for 6% implied growth against ~10%
expected (composite lead 0.7, outside the 0.5 tie band). Why not QRTZ: a better business by a little, priced for a lot. Flip condition:
QRTZ below a hypothetical $x (the price at which its composite reaches 7.0), or ZEPH's top
customer moving above 25% of revenue. Diversification: ZEPH/QRTZ weekly correlation 0.83,
same top driver (enterprise IT budgets) → own one, not both. MNTL: fastest grower, priced
for it, and the highest dilution — third.

Signal for the pick: ZEPH — composite 7.0 → BULLISH, MODERATE conviction, LONG-TERM,
Confidence MEDIUM (consensus growth figures pasted, not verified). Next: stock-eval,
stock-valuation and bear-case on ZEPH before any order.

Notes

  • Same date, same window, same accounting. Most bad comparisons are bad data alignment, not bad judgement. Calendarize fiscal years and expense SBC before scoring anything.
  • Anchors are absolute on purpose. A within-group rank would always crown someone a 10; absolute anchors let "none of these is attractive" be the honest answer.
  • The composite is evidence for the argument, not the argument. Two composites within 0.5 are a tie; a 0.3-point lead is never "the reason".
  • Cross-sector comparisons are allowed but flagged. A bank against a software company can be argued (on the user's objective) but not scored on one composite — Phase 1's warning is mandatory.
  • Five tickers maximum. Beyond that the argument collapses into a ranking; use stock-screener.
  • Before: stock-screener to shortlist; competitor-analysis for moat ratings. After: stock-eval, stock-valuation, bear-case on the winner; thesis-tracker to write it down; position-ladder for sizing; risk-stress-test if more than one is owned.

Thesis Invalidation

After delivering the analysis signal, specify what would reverse it:

If signal is BULLISH (the pick is a buy) — thesis breaks if:

  • The flip condition from Phase 4 is met — the runner-up's price falls to the flip price, or the winner's deciding KPI crosses the stated level
  • The winner's deciding dimension deteriorates by more than 2 sub-score points at the next filing (e.g. its valuation sub-score falls because the price ran, not the business improved — the edge has been paid out)
  • An explanation for the Phase 6 valuation gap surfaces (a lawsuit, an accounting restatement, a customer loss) — the cheap stock was correctly cheap

If signal is BEARISH (the whole group is unattractive) — thesis breaks if:

  • A group-wide de-rating moves the best valuation sub-score up by 3+ points with quality and growth unchanged — the businesses were fine, the prices were the problem
  • The shared driver named in Phase 4 step 5 (rates, a capex cycle, a regulation) turns favourable with evidence in reported numbers, not forecasts
  • A new candidate with the same exposure scores above 6.0 — add it and re-run rather than settle for the old group

Re-run this analysis when:

  • Next earnings release of any ticker in the comparison
  • Price moves ±15% from current level for any ticker (valuation sub-scores move with price)
  • 60 days have elapsed
  • Material news event (acquisition, leadership change, regulatory decision)

Standard Signal Output

The box expresses the view on the winning pick only, on an absolute basis (Phase 7) — not on the group and not on the runner-up. Name the winning ticker in the line directly above the box ("Signal for the pick: "). Signal: BULLISH means the winner is attractive on its own merits; NEUTRAL means it is the best of the group but not a buy by itself (or the group was not comparable); BEARISH means none of the candidates is attractive at current prices — Action: SELL then means "do not buy any of these; trim if already owned", never "short the group".

All analysis concludes with this standardized block:

In the output, the box below is followed immediately by this machine-readable footer — the same values as the box, filled in (no placeholders, no comments), so tools can read the result without parsing the box:

{"investskill": "1", "skill": "compare", "ticker": "<TICKER, or null for a portfolio / market-wide view>", "as_of": "<YYYY-MM-DD from Data & Sources>", "signal": "<BULLISH | NEUTRAL | BEARISH>", "confidence": "<HIGH | MEDIUM | LOW>", "horizon": "<SHORT-TERM | MEDIUM-TERM | LONG-TERM>", "score": 0.0, "action": "<BUY | HOLD | SELL>", "conviction": "<STRONG | MODERATE | WEAK>", "retrieval": "<pasted by user | web/tool retrieval | model memory>", "sources": ["<each primary source, dated>"]}
╔══════════════════════════════════════════════╗
║              INVESTMENT SIGNAL               ║
╠══════════════════════════════════════════════╣
║ Signal:      BULLISH / NEUTRAL / BEARISH     ║
║ Confidence:  HIGH / MEDIUM / LOW             ║
║ Horizon:     SHORT / MEDIUM / LONG-TERM      ║
║ Score:       X.X / 10                        ║
╠══════════════════════════════════════════════╣
║ Action:      BUY / HOLD / SELL               ║
║ Conviction:  STRONG / MODERATE / WEAK        ║
╚══════════════════════════════════════════════╝

Score Guide: 8.0–10.0 Strongly Bullish | 6.0–7.9 Moderately Bullish | 4.0–5.9 Neutral | 2.0–3.9 Moderately Bearish | 0.0–1.9 Strongly Bearish Confidence: HIGH (strong data, clear signals) | MEDIUM (mixed signals) | LOW (limited data, conflicting signals) Horizon: SHORT-TERM (1 week–3 months) | MEDIUM-TERM (3 months–1 year) | LONG-TERM (1+ years)

Note: The Score in the box is the winning pick's composite from Phase 3 (adjusted per Phase 7), so it is comparable with a stock-eval score on the same ticker; the other tickers' composites are reported in the head-to-head table, not in the box.

Disclaimer: Educational analysis only. Not financial advice.

Competitor Analysis

⚠️ Data Verification — Do This Before Any Analysis

Before running any analysis, always retrieve the latest market data for the ticker:

  1. Fetch current price — use web search or ask the user for the live price, 52-week range, and market cap. Never assume a price from training data.
  2. Confirm key figures — revenue, segment mix, gross and operating margin, ROIC and market share for the company and each rival from the same period (latest fiscal year or TTM), so the comparison is like for like.
  3. State your data source — fill in the Data & Sources header (next section) so the origin, as-of date, retrieval path, and confidence of every figure are explicit at the top of the output.
  4. Flag stale data explicitly — if live data is unavailable, display this warning before proceeding:

⚠️ Live data unavailable. The following analysis uses training-data estimates which may be significantly out of date. Verify all prices and metrics before making any decisions.

Never silently substitute training-data estimates for current prices. When in doubt, ask the user to paste the latest quote.


📋 Data & Sources Header — Open Every Output With It

The first thing in the output is this provenance block, filled in — never left as placeholders. It is the standard documented on the Data & Accuracy page and the first thing result-validator looks for:

Data & Sources
  As of:      <date the figures represent, e.g. 2026-06-30>
  Source:     <primary docs — SEC EDGAR 10-K/10-Q, company IR, FRED, exchange data …>
  Retrieval:  <pasted by user | web/tool retrieval | model memory>
  Confidence: <HIGH | MEDIUM | LOW>
  • Retrieval: model memory must be paired with Confidence: LOW — memory is a placeholder until confirmed against a primary source.
  • Mixed sources: list each with its own as-of date rather than blending them.
  • Sources disagree on the same figure (a closing price, a deal term): prefer the primary source (SEC filing, company release, exchange data) over an aggregator, then the most recent dated one. Show both values with their sources and dates, say which you used and why, and never average them.
  • Data the user pasted is reported as pasted by user; do not upgrade its confidence beyond what the user's own source supports.

🌐 Output Language — --lang zh-TW

English is the default. With --lang zh-TW — or when the user writes in Traditional Chinese and asks for a Chinese answer — write the whole analysis in Traditional Chinese (繁體中文, Taiwan usage), keeping it machine-checkable:

  • Translate every heading, narrative paragraph, table header and bullet. Give each metric's English name in brackets the first time it appears — 自由現金流(Free Cash Flow, FCF) — so the reader can match it to the glossary and to English analyses.
  • Keep in English tickers, figures and units, SEC form and item names (10-K Item 7, Form 4), the four Data & Sources field labels (As of, Source, Retrieval, Confidence — the Retrieval value may be written in Chinese, e.g. 使用者提供), and the Investment Signal box and JSON footer exactly as specified — field names and the BULLISH / NEUTRAL / BEARISH, HIGH / MEDIUM / LOW, BUY / HOLD / SELL, STRONG / MODERATE / WEAK values. Tools parse them.
  • Title the invalidation section 論點失效條件(Thesis Invalidation), and close with the disclaimer in Chinese: 免責聲明: 僅供教育用途,非投資建議。
  • Use Taiwan financial terms: 殖利率、本益比、營業利益率、自由現金流、聯準會、選擇權、買權/賣權、除息日、預扣稅.

Conduct deep competitive moat analysis to assess whether a company has a durable competitive advantage, how wide that moat is, and what the competitive dynamics of its industry mean for long-term investment returns.

Overview

Competitive analysis answers the fundamental question: "Does this company have a durable competitive advantage, and how wide is its moat?" This directly determines the appropriate valuation premium or discount vs. the sector.

A company with a wide, widening moat deserves a premium P/E and P/FCF multiple because its excess returns on capital are durable. A company with no moat, or a narrowing moat, should trade at or below sector multiples regardless of near-term earnings momentum. Understanding the moat is the single most important determinant of a stock's long-term investment return — more important than any individual quarterly earnings figure.

This skill provides a structured, repeatable framework for moat identification, industry attractiveness scoring, competitive benchmarking, and innovation positioning. Output feeds directly into stock-valuation (to set appropriate WACC and terminal growth rate) and stock-eval (to contextualize ROIC and margin trends). To weigh this company against two to four rivals as investments rather than as businesses, use compare.


1. Moat Identification Framework

Five Sources of Economic Moat (Morningstar Framework)

1. Network Effects Value increases with each additional user or participant in the platform or network.

  • Examples: Visa, Mastercard (payment networks), Meta (social graph), Airbnb (marketplace), Microsoft 365 (collaboration network)
  • Test: Does adding users benefit existing users? Does the network become more valuable as it scales?
  • Signs of network effects: Organic user growth with low CAC, high retention as network grows, winner-take-most dynamics

2. Cost Advantages Structural ability to produce goods or services at lower cost than competitors.

  • Sources: Scale advantages (fixed cost leverage), process innovation (proprietary manufacturing), geographic advantage (proximity to inputs), unique asset access (mining rights, owned infrastructure)
  • Examples: Costco (buying scale + lean operations), Amazon (logistics scale), Nucor (mini-mill process innovation)
  • Test: Can competitors replicate this cost structure at competitive cost? What would it cost to build?

3. Intangible Assets Brands, patents, regulatory licenses, or proprietary data that competitors cannot easily copy.

  • Brand: Can the company charge a premium price based on brand perception alone? (Apple, LVMH, Coca-Cola)
  • Patents: How many years of exclusivity remain? Is the IP portfolio broad or narrow?
  • Regulatory licenses: Are licenses scarce, non-transferable, or expensive to obtain? (broadcast licenses, pharmaceutical approvals, financial licenses)
  • Proprietary data: Is the data asset self-reinforcing and competitively irreplicable?
  • Test: Can the company charge premium prices, or does the intangible give exclusive market access?

4. Switching Costs High cost — financial, operational, or psychological — for customers to change providers.

  • Financial switching costs: Contractual lock-in, migration costs, retraining costs
  • Operational switching costs: Deep workflow integration, data portability limitations
  • Psychological switching costs: Brand loyalty, habit formation, trust
  • Examples: Oracle (ERP deeply embedded in operations), Salesforce (CRM data and workflow integration), Adobe (creative suite skill investment)
  • Test: What % of customers have churned in the last 3 years? How long is the average customer tenure?

5. Efficient Scale Company operates in a market that can profitably support only one or a few competitors, creating natural oligopoly or monopoly dynamics.

  • Examples: Waste Management (regional landfills), utility companies, specialty chemicals with natural regional monopolies
  • Test: Would a new entrant earn below-cost returns given the existing market structure? Is the total addressable market too small to profitably split further?

Moat Width Assessment

Moat Width    Definition                                 ROIC Signal
──────────────────────────────────────────────────────────────────────
Wide Moat     Clear, sustainable advantage 20+ years     ROIC consistently and
              Structural barriers that are durable        significantly > WACC
              Premium P/E valuation appropriate

Narrow Moat   Some advantages, 10–20 year durability      ROIC modestly > WACC
              Barriers exist but can be overcome           Slight premium warranted
              with sufficient capital or time

No Moat       No sustainable competitive advantage         ROIC ≈ WACC or below
              Commodity-like pricing dynamics              In-line sector valuation
              New entrants can replicate economics

Moat at Risk  Previously existing moat is eroding          ROIC declining toward
              Structural disruption underway                or below WACC
              Discount valuation warranted

Moat Trend (most important forward-looking question):

  • Widening: Competitive advantages are strengthening, market share is growing, ROIC is increasing
  • Stable: Moat is intact but not materially widening; returns on capital are consistent
  • Narrowing: Competitive pressure, disruption, or commoditization is compressing the moat

2. Porter's Five Forces Deep Analysis

Force 1: Competitive Rivalry (Intensity within industry)

How intensely do existing competitors compete for market share?

  • Number and size distribution of competitors (fragmented vs. concentrated)
  • Industry growth rate (slow-growth industries intensify rivalry; fast-growing markets reduce it)
  • Product differentiation level (commodity products = intense price competition)
  • Exit barriers (high exit barriers trap capacity in the market, intensifying rivalry)
  • Fixed cost intensity (high fixed costs create pressure to fill capacity at any price)

Rivalry Intensity: Low / Moderate / High / Extreme

Force 2: Threat of New Entrants

How easily can new competitors enter the market?

  • Capital requirements for market entry (low capital = easier entry)
  • Economies of scale advantages for incumbents
  • Network effect barriers (winner-take-most dynamics deter entry)
  • Regulatory and licensing barriers (FDA approvals, financial licenses, environmental permits)
  • Brand and customer loyalty barriers (how long would it take to build credibility?)
  • Access to distribution channels
  • Incumbent cost advantages independent of scale (patents, proprietary processes)

Entry Threat: Low / Moderate / High

Force 3: Bargaining Power of Suppliers

How much leverage do input suppliers have over the company?

  • Supplier concentration vs. buyer concentration (few suppliers, many buyers = high supplier power)
  • Uniqueness and criticality of the supplied product or service
  • Cost of switching suppliers (sole-source vs. multi-source supply chains)
  • Supplier forward integration threat (can suppliers bypass the company and sell direct?)
  • Importance of the company to the supplier's revenue (are you a large or small customer?)

Supplier Power: Low / Moderate / High

Force 4: Bargaining Power of Buyers (Customers)

How much leverage do customers have to negotiate price or terms?

  • Customer concentration (what % of revenue comes from the top 10 customers?)
  • Price sensitivity of customers (is the purchase a major budget item or negligible?)
  • Switching costs for customers (low switching costs = high buyer power)
  • Buyer backward integration threat (can customers build this capability in-house?)
  • Availability of information (informed buyers negotiate better)
  • Volume buying leverage (large customers extract better terms)

Buyer Power: Low / Moderate / High

Force 5: Threat of Substitutes

What alternatives exist outside the direct competitive set?

  • Availability of substitute products or services (different product, same customer job-to-be-done)
  • Price-performance improvement rate of substitutes (is the substitute improving faster than the incumbent?)
  • Customer propensity to substitute (how much switching actually happens?)
  • Relative price of substitutes (cheap substitute + acceptable quality = high threat)
  • Example: Streaming vs. cable TV; cloud computing vs. on-premise hardware; electric vehicles vs. internal combustion

Substitute Threat: Low / Moderate / High

Five Forces Summary Score

Score each force 1–5, where 5 = most favorable for the company being analyzed:

Force                      Score (1-5)    Assessment
─────────────────────────────────────────────────────────────────
Competitive Rivalry         [1-5]         [description of key dynamics]
New Entrant Threat          [1-5]         [key barriers or lack thereof]
Supplier Power              [1-5]         [key supplier dynamics]
Buyer Power                 [1-5]         [customer concentration, switching costs]
Substitute Threat           [1-5]         [main substitutes and their threat level]
─────────────────────────────────────────────────────────────────
Industry Attractiveness Score:    [X.X / 5]

Score Interpretation:
4.5–5.0  Extremely attractive industry (structural advantages strong)
3.5–4.4  Attractive industry (mostly favorable dynamics)
2.5–3.4  Average industry (mixed dynamics)
1.5–2.4  Unattractive industry (structural headwinds)
1.0–1.4  Highly unattractive (commodity, intense competition, low returns)

3. Market Share Analysis

Understand whether the company is gaining, maintaining, or losing ground in its market:

  • Current market share % and 3-year trend (gaining / stable / losing)
  • Market share concentration (Herfindahl-Hirschman Index — HHI — of the industry)
    • HHI > 2,500: Highly concentrated (oligopoly/monopoly dynamics)
    • HHI 1,500–2,500: Moderately concentrated
    • HHI < 1,500: Fragmented (competitive)
  • Market share growth mechanics: Organic capture vs. M&A-driven share; price-led vs. volume-led
  • Geographic market share variations: May be dominant in home market, subscale internationally, or vice versa
  • Segment market share: Company may have commanding share in a high-value niche while being subscale in commodity segments
  • Share gain velocity: Rate of change matters as much as absolute level. A company gaining 0.5% share annually in a large market is a powerful signal.

4. Competitive Benchmarking

Compare the company vs. its top 3–5 direct competitors across key financial and operational metrics:

Metric                [Company]   [Comp 1]   [Comp 2]   [Comp 3]   Industry Avg
───────────────────────────────────────────────────────────────────────────────────
Revenue Growth (3yr)  [%]         [%]        [%]        [%]        [%]
Gross Margin          [%]         [%]        [%]        [%]        [%]
Operating Margin      [%]         [%]        [%]        [%]        [%]
Net Margin            [%]         [%]        [%]        [%]        [%]
ROIC                  [%]         [%]        [%]        [%]        [%]
ROE                   [%]         [%]        [%]        [%]        [%]
Revenue per Employee  [$k]        [$k]       [$k]       [$k]       [$k]
Customer Retention    [%]         [%]        [%]        [%]        [%]
R&D as % Revenue      [%]         [%]        [%]        [%]        [%]
Gross Profit per $R&D [ratio]     [ratio]    [ratio]    [ratio]    [ratio]
NPS Score             [score]     [score]    [score]    [score]    [score]
Market Share %        [%]         [%]        [%]        [%]        —

Interpretation: Highlight where the company leads, lags, or matches the peer set. ROIC > industry average consistently = moat evidence. Gross margin premium = pricing power or cost advantage. Higher revenue per employee = efficiency advantage.


5. Innovation & Disruption Assessment

Evaluate whether the company is positioned as a disruptor or a potential target of disruption:

  • R&D investment level and productivity:
    • R&D as % of revenue (spending level)
    • Patents filed per $1M R&D (output efficiency)
    • Time-to-market for new product launches vs. peers
  • Product roadmap visibility: Does management articulate a clear multi-year innovation roadmap with specific milestones?
  • Technology platform assessment: Is the core technology platform modern (cloud-native, API-first, modular) or legacy (monolithic, on-premise, technical debt-laden)?
  • Disruption positioning: Is this company the disruptor or the disrupted?
    • Disruptor indicators: Taking share from incumbents, serving underserved segments, improving price-performance faster than industry
    • Disrupted indicators: Losing share to newer platforms, customers migrating to substitutes, pricing power declining
  • Adjacent market opportunities: What is the total addressable market (TAM) expansion potential? Can the moat extend into adjacent categories?
  • AI/software/platform disruption threat: Is the industry undergoing a platform shift that could rapidly alter competitive dynamics? (e.g., AI replacing workflow software, direct-to-consumer bypass of distributors)

6. Management Quality in Competitive Context

Assess whether management is executing effectively in the competitive environment:

  • Capital allocation track record: Has management invested capital at returns above WACC? What is the M&A track record (value-creative or value-destructive)?
  • Competitive response speed: How quickly does management respond to competitive threats? (pricing changes, product updates, strategic pivots)
  • Innovation culture indicators: Employee Glassdoor ratings vs. competitors; pace of product launches; engineering talent density (LinkedIn data); Blind/levels.fyi compensation vs. peers
  • CEO competitive vision: How does the CEO discuss competition in earnings calls? Dismissive, realistic, or strategically insightful?
  • Track record vs. stated strategy: Has management delivered on prior competitive strategy commitments? Or does strategy change frequently without execution?

7. Pricing Power Analysis

Quantify the company's ability to raise prices without losing customers:

  • Premium vs. discount pricing: Does the company price above, at, or below competitors? What is the quantified price premium?
  • Price increase history: Has the company raised prices in the last 5 years? Did volume decline, remain stable, or grow despite price increases? (volume stability after price increases = strong pricing power)
  • Customer willingness-to-pay research: NPS scores, customer satisfaction surveys, retention data, and churn analysis provide indirect evidence of willingness to pay
  • Gross margin expansion/compression trend: Expanding gross margins while growing revenue = pricing power. Compressing gross margins under competitive pressure = pricing power erosion.
  • Price elasticity indicators: For consumer businesses, track promotional intensity. Excessive discounting = inability to hold price. For B2B, track deal cycle length and discount rates.

8. Moat Score Composite

Moat Scorecard:
Component                    Weight    Score (0-10)    Notes
──────────────────────────────────────────────────────────────────────
Moat Source Strength           25%      [0-10]         [which of 5 sources are present]
Moat Durability (years)        20%      [0-10]         [wide/narrow/none, estimated longevity]
Competitive Position           20%      [0-10]         [gaining/stable/losing vs. peers]
Industry Attractiveness        15%      [0-10]         [Five Forces score converted to 0-10]
Pricing Power                  10%      [0-10]         [premium pricing, margin trend]
Innovation Positioning         10%      [0-10]         [disruptor/neutral/disrupted]
──────────────────────────────────────────────────────────────────────
Composite Moat Score:         100%      X.X / 10

Moat Assessment:
8–10: Wide Moat (significant valuation premium justified; durable excess returns)
6–8:  Narrow Moat (modest premium warranted; monitor for narrowing)
4–6:  No Clear Moat (in-line with sector valuation; commodity-like returns)
0–4:  Moat at Risk (valuation discount warranted; sell consideration)

Scoring Reference:

  • Moat Source Strength: 9–10 = 3+ strong, reinforcing moat sources; 7–8 = 2 clear sources; 5–6 = 1 credible source; 3–4 = partial/debatable source; 0–2 = no identifiable moat
  • Moat Durability: 9–10 = 20+ year visibility; 7–8 = 15+ years; 5–6 = 10 years; 3–4 = 5 years; 0–2 = structural disruption underway
  • Industry Attractiveness: Derived from Five Forces score (0–5 scale) × 2 to convert to 0–10

9. Competitive Intelligence Sources

Use these primary and secondary research sources to build the competitive picture:

Regulatory and Financial Filings:

  • SEC 10-K "Business" and "Competition" sections — required disclosure of competitive dynamics
  • SEC 10-K "Risk Factors" — management's own description of competitive threats
  • DEF 14A (proxy statement) — executive compensation tied to competitive metrics
  • Competitor 10-K filings — cross-reference to understand industry dynamics

Management and Qualitative Sources:

  • Earnings call transcripts — frequency and language around competitors signals competitive intensity
  • Investor Day presentations — long-term competitive strategy and management conviction
  • Glassdoor and LinkedIn — employee satisfaction vs. competitors, talent flow analysis

Customer and Market Research:

  • G2 / Capterra / TrustRadius — B2B software competitive ratings and reviews
  • Yelp / Google Reviews — consumer business competitive positioning
  • JD Power — automotive and consumer product competitive rankings
  • NPS benchmarks by industry (Bain & Company publishes)

Technology and Innovation Intelligence:

  • Google Patents / USPTO — patent portfolio analysis and competitive IP positioning
  • Job postings analysis (LinkedIn, Indeed) — what skills a company is hiring for signals its strategic direction
  • GitHub (for software companies) — open-source activity and developer ecosystem strength
  • AlternativeTo / ProductHunt — consumer product competitive landscape mapping

Industry Research:

  • Gartner Magic Quadrant and Critical Capabilities reports
  • Forrester Wave reports
  • IDC market share data
  • Trade publications specific to the industry vertical

10. Input Formats

# Single company analysis
/competitor-analysis AAPL

# With specific sector context for more targeted analysis
/competitor-analysis MSFT --sector "cloud computing"

# With specified peer group for benchmarking
/competitor-analysis NVDA --peers AMD,INTC,QCOM

# Moat analysis only (faster, focused output)
/competitor-analysis GOOGL --moat-only

# Full analysis with visual output for /report-generator
/competitor-analysis AMZN --visual

11. Visualization Support

When --visual flag is used, include chart data tables for report generation:

Porter's Five Forces Radar Chart

Chart Type: Radar/spider chart

Force                   Score (1-5)
Competitive Rivalry     [value]
New Entrant Threat      [value]
Supplier Power          [value]
Buyer Power             [value]
Substitute Threat       [value]

Competitive Benchmarking Chart

Chart Type: Grouped bar chart

Metric          [Company]   [Comp 1]   [Comp 2]   [Comp 3]   Industry Avg
Gross Margin    [%]         [%]        [%]        [%]        [%]
ROIC            [%]         [%]        [%]        [%]        [%]
Revenue Growth  [%]         [%]        [%]        [%]        [%]

Moat Score Components Bar Chart

Chart Type: Horizontal bar chart with weighted scores

Component               Weighted Score
Moat Source Strength    [value × 0.25]
Moat Durability         [value × 0.20]
Competitive Position    [value × 0.20]
Industry Attractiveness [value × 0.15]
Pricing Power           [value × 0.10]
Innovation Positioning  [value × 0.10]
─────────────────────────────────────
Composite Moat Score:   [X.X / 10]

Output

Provide a comprehensive competitive analysis report with:

  • Executive Summary (moat assessment, competitive position, trend — 3 sentences)
  • Moat Identification (sources present, width, trend, durability estimate)
  • Porter's Five Forces Analysis with individual force scores and industry attractiveness score
  • Market Share Analysis and 3-year trend
  • Competitive Benchmarking table vs. top 3–5 peers
  • Innovation & Disruption Assessment
  • Pricing Power Analysis
  • Management Competitive Quality
  • Composite Moat Score Card
  • Investment Implications (how moat assessment affects valuation, key competitive risks, bull/bear case for competitive position)

Enhanced Output (with --visual flag)

  • All standard sections above
  • Porter's Five Forces radar chart data
  • Competitive benchmarking grouped bar chart data
  • Moat score components horizontal bar chart data
  • Chart specifications for HTML report generation via /report-generator

Thesis Invalidation

After delivering the analysis signal, specify what would reverse it:

If signal is BULLISH — thesis breaks if:

  • well-funded competitor enters core market OR key customer (>15% revenue) lost
  • [One or two more triggers drawn from this analysis's own drivers, each with a threshold]

If signal is BEARISH — thesis breaks if:

  • moat-widening acquisition announced OR key competitor files bankruptcy
  • [One or two more triggers drawn from this analysis's own drivers, each with a threshold]

Re-run this analysis when:

  • Next earnings release
  • Price moves ±15% from current level
  • 60 days have elapsed
  • Material news event (acquisition, leadership change, regulatory decision)

Standard Signal Output

All analysis concludes with this standardized block:

In the output, the box below is followed immediately by this machine-readable footer — the same values as the box, filled in (no placeholders, no comments), so tools can read the result without parsing the box:

{"investskill": "1", "skill": "competitor-analysis", "ticker": "<TICKER, or null for a portfolio / market-wide view>", "as_of": "<YYYY-MM-DD from Data & Sources>", "signal": "<BULLISH | NEUTRAL | BEARISH>", "confidence": "<HIGH | MEDIUM | LOW>", "horizon": "<SHORT-TERM | MEDIUM-TERM | LONG-TERM>", "score": 0.0, "action": "<BUY | HOLD | SELL>", "conviction": "<STRONG | MODERATE | WEAK>", "retrieval": "<pasted by user | web/tool retrieval | model memory>", "sources": ["<each primary source, dated>"]}
╔══════════════════════════════════════════════╗
║              INVESTMENT SIGNAL               ║
╠══════════════════════════════════════════════╣
║ Signal:      BULLISH / NEUTRAL / BEARISH     ║
║ Confidence:  HIGH / MEDIUM / LOW             ║
║ Horizon:     SHORT / MEDIUM / LONG-TERM      ║
║ Score:       X.X / 10                        ║
╠══════════════════════════════════════════════╣
║ Action:      BUY / HOLD / SELL               ║
║ Conviction:  STRONG / MODERATE / WEAK        ║
╚══════════════════════════════════════════════╝

Score Guide: 8.0–10.0 Strongly Bullish | 6.0–7.9 Moderately Bullish | 4.0–5.9 Neutral | 2.0–3.9 Moderately Bearish | 0.0–1.9 Strongly Bearish Confidence: HIGH (strong data, clear signals) | MEDIUM (mixed signals) | LOW (limited data, conflicting signals) Horizon: SHORT-TERM (1 week–3 months) | MEDIUM-TERM (3 months–1 year) | LONG-TERM (1+ years)

Disclaimer: Educational analysis only. Not financial advice.

Credit Check — Can the Equity Survive the Debt?

⚠️ Data Verification — Do This Before Any Analysis

Before running any analysis, always retrieve the latest market data for the ticker:

  1. Fetch current price — use web search or ask the user for the live price, 52-week range, and market cap. Never assume a price from training data.
  2. Pull the debt disclosures from the filings, not a data feed — aggregator "total debt" figures routinely omit leases, mix in or leave out current maturities, and lag refinancings. Retrieve:
    • 10-K / latest 10-Q debt footnote — every instrument: principal, coupon (fixed or floating and its reference rate), maturity, security/seniority, and the scheduled maturities table (next five years and thereafter).
    • MD&A "Liquidity and Capital Resources" — cash, undrawn revolver capacity and its expiry, commercial paper, planned refinancing, management's stated liquidity needs; and the contractual obligations discussion.
    • Credit agreement / indenture (8-K Item 1.01 exhibits, 10-K Exhibit 10.x / 4.x) — financial covenants (maximum leverage, minimum coverage, springing tests) and their definitions of EBITDA; the 10-Q covenant-compliance statement.
    • Lease note (ASC 842 operating and finance lease liabilities), pension and OPEB note (funded status), commitments and contingencies (guarantees, purchase obligations), and any supplier-finance / receivables-factoring disclosure.
    • Convertible notes and preferred stock terms (conversion price, capped calls, put dates).
    • Credit ratings and outlooks (S&P, Moody's, Fitch press releases or the company's IR page) and, if available, bond prices / yields (FINRA TRACE) or CDS spreads; benchmark yields by rating from FRED (ICE BofA indices, e.g. BAMLC0A4CBBB OAS and BAMLC0A4CBBBEY effective yield for BBB; BAMLH0A1HYBB, BAMLH0A2HYB, BAMLH0A3HYC for BB / B / CCC & lower). Any 8-K after the latest 10-Q (new notes, tender offers, revolver amendments) supersedes the footnote — check for them.
  3. State your data source — fill in the Data & Sources header (next section) so the origin, as-of date, retrieval path, and confidence of every figure are explicit at the top of the output.
  4. Flag stale data explicitly — if live data is unavailable, display this warning before proceeding:

⚠️ Live data unavailable. The following analysis uses training-data estimates which may be significantly out of date. Verify all prices and metrics before making any decisions.

Never silently substitute training-data estimates for current prices. When in doubt, ask the user to paste the latest quote.


📋 Data & Sources Header — Open Every Output With It

The first thing in the output is this provenance block, filled in — never left as placeholders. It is the standard documented on the Data & Accuracy page and the first thing result-validator looks for:

Data & Sources
  As of:      <date the figures represent, e.g. 2026-06-30>
  Source:     <primary docs — SEC EDGAR 10-K/10-Q, company IR, FRED, exchange data …>
  Retrieval:  <pasted by user | web/tool retrieval | model memory>
  Confidence: <HIGH | MEDIUM | LOW>
  • Retrieval: model memory must be paired with Confidence: LOW — memory is a placeholder until confirmed against a primary source.
  • Mixed sources: list each with its own as-of date rather than blending them.
  • Sources disagree on the same figure (a closing price, a deal term): prefer the primary source (SEC filing, company release, exchange data) over an aggregator, then the most recent dated one. Show both values with their sources and dates, say which you used and why, and never average them.
  • Data the user pasted is reported as pasted by user; do not upgrade its confidence beyond what the user's own source supports.

🌐 Output Language — --lang zh-TW

English is the default. With --lang zh-TW — or when the user writes in Traditional Chinese and asks for a Chinese answer — write the whole analysis in Traditional Chinese (繁體中文, Taiwan usage), keeping it machine-checkable:

  • Translate every heading, narrative paragraph, table header and bullet. Give each metric's English name in brackets the first time it appears — 自由現金流(Free Cash Flow, FCF) — so the reader can match it to the glossary and to English analyses.
  • Keep in English tickers, figures and units, SEC form and item names (10-K Item 7, Form 4), the four Data & Sources field labels (As of, Source, Retrieval, Confidence — the Retrieval value may be written in Chinese, e.g. 使用者提供), and the Investment Signal box and JSON footer exactly as specified — field names and the BULLISH / NEUTRAL / BEARISH, HIGH / MEDIUM / LOW, BUY / HOLD / SELL, STRONG / MODERATE / WEAK values. Tools parse them.
  • Title the invalidation section 論點失效條件(Thesis Invalidation), and close with the disclaimer in Chinese: 免責聲明: 僅供教育用途,非投資建議。
  • Use Taiwan financial terms: 殖利率、本益比、營業利益率、自由現金流、聯準會、選擇權、買權/賣權、除息日、預扣稅.

Overview

Equity sits last in the capital structure. Lenders get paid first, set covenants, and — when a company cannot refinance — take the business. Most equity frameworks in this catalog treat debt as a line in the balance-sheet section or a number in the WACC; none asks the creditor's question from the shareholder's seat: when does the debt come due, what will it cost to roll at today's rates, can the company pay it without diluting or wiping out the equity, and what does the credit market already think?

This skill reads the balance sheet the way a credit analyst does, then translates it back to the equity. It builds the maturity wall against cash and free cash flow, prices the refinancing shock at current yields for the company's rating, tracks coverage and leverage against approximate rating bands and the company's own covenants, finds the debt that is not called debt (leases, pensions, guarantees, supplier finance), and ends with an equity-holder verdict: is the stock the residual claim on a healthy business, or an option on survival? The headline number is a Credit Health Score (0–10, higher = healthier).

Where it sits. financial-report-analyst and stock-eval touch leverage in passing; bear-case argues a refinancing thesis without the schedule; risk-stress-test shocks prices, not balance sheets. Run this skill when leverage is the swing factor, then hand the refinancing scenario to bear-case, the liquidity runway to risk-stress-test, and the dividend-coverage question to dividend-analysis. If the Altman Z reading in forensic-accounting is grey or distressed, this is the next skill to run.


1. Inputs

InputRequiredDefault if unstated
Ticker✅—
Filings pasted by the user (debt footnote, MD&A liquidity, credit agreement)optionalretrieve from SEC EDGAR; list accession numbers
Base rates for the refinancing shock (Treasury curve, rating-band yields)optionalfetch from FRED; state the date
Rating to use when the company is unratedoptionalinfer an implied band from the table in Phase 6 and label it "implied"
Stress case for EBITDAoptional−20% EBITDA (−40% for cyclicals), stated
Holding horizonoptional1–3 years (covers most maturity walls)

Pre-profit mode. If EBITDA is negative or the company burns cash, coverage and leverage ratios are meaningless — the analysis switches to liquidity runway (Phase 4) and dilution history, and the score weights change (Phase 8).


2. Framework

Phase 1 — Debt schedule and maturity wall

Build one table from the debt footnote, one row per instrument:

InstrumentPrincipal ($M)CouponFixed / Floating (reference rate + spread)MaturitySecured / UnsecuredCallable / PutableNotes

Then the maturity wall — cumulative maturities against the resources available to meet them:

WindowMaturing debt (cumulative)Cash & ST investmentsUndrawn committed revolver (expiring after the window)Cumulative FCF (base case)Coverage ratio
Next 12 months
Next 2 years
Next 3 years
Next 5 years
Coverage ratio (window) = (Cash + Undrawn revolver + Cumulative FCF) / Cumulative maturities
Coverage ratioReading
> 2.0×Self-funded — maturities can be repaid without the capital markets
1.0× – 2.0×Manageable — refinancing preferred but not required
0.5× – 1.0×Refinancing-dependent — the company needs open markets in that window
< 0.5×Wall — a closed market or a downgrade in that window is an equity event

A revolver that expires before the maturity it is meant to backstop does not count. Note springing maturities (a term loan that accelerates if a later bond is not refinanced) — they move the wall forward.

Fixed vs. floating mix. Report % floating (after disclosed swaps). Each 100 bp move in the reference rate changes interest expense by floating principal × 1% — show it in $ and as % of EBIT.

Phase 2 — Cost of debt and the refinancing shock

Weighted average coupon (WAC) = Σ (principal_i × coupon_i) / Σ principal_i
Market yield                  = today's yield for the company's rating and a similar tenor
                                (the company's own bonds via TRACE if traded; otherwise the ICE BofA rating-band effective yield from FRED)

For each tranche maturing in the next three years:

ΔInterest_i       = Principal_i × (Market yield − Coupon_i)
Refinancing shock = Σ ΔInterest_i               ($ per year, once all are rolled)
Shock % of EBIT   = Refinancing shock / EBIT (LTM)
EPS impact        = Refinancing shock × (1 − tax rate) / Diluted shares
Shock % of EBITReading
< 5%Immaterial
5% – 15%A visible earnings headwind — check whether consensus includes it
> 15%Material — the refinancing alone changes the investment case

Debt issued during the 2020–2021 low-rate window often carries coupons far below today's yields; name the cheapest tranches and when they roll. A negative shock (rates below coupons) is a tailwind — say so.

Phase 3 — Coverage, leverage and cash-flow ratios (5-year trend)

MetricFormulaTrend (5 yrs)Stress case
EBIT / interestEBIT / Interest expense
EBITDA / interestEBITDA / Interest expense
Debt / EBITDATotal debt / EBITDA
Net debt / EBITDA(Total debt − Cash & ST investments) / EBITDA
Adjusted debt / EBITDAR(Debt + Lease liabilities + After-tax pension deficit) / (EBITDA + Rent)
FFO / debt(Net income + D&A + Deferred taxes + other non-cash items) / Total debt — or (EBITDA − Cash interest − Cash taxes) / Debt, S&P-style
FCF / debt(OCF − Capex) / Total debt
Debt / capitalTotal debt / (Total debt + Book equity)

Rules. Use GAAP EBITDA as the base and show management's "Adjusted EBITDA" alongside — adjustments (synergies, add-backs, SBC) can flatter leverage by a turn or more. Use interest expense plus capitalised interest. The stress case applies the stated EBITDA haircut with debt held constant. Direction matters as much as level: coverage falling for three years while leverage rises is the pattern that precedes downgrades.

Phase 4 — Liquidity runway (pre-profit and cash-burning companies)

Liquidity       = Cash & ST investments + Undrawn committed revolver (net of covenant-restricted capacity)
Quarterly burn  = −(OCF − Capex), average of the last 4 quarters (use the latest quarter if burn is accelerating)
Runway (quarters) = Liquidity / Quarterly burn
RunwayReading
> 12 quartersFunded through most plans
6 – 12 quartersExpect a raise within the horizon — price in dilution
4 – 6 quartersRaise imminent — the timing is the company's weakest negotiating moment
< 4 quartersSurvival question — going-concern language likely

Report dilution history (shares outstanding over 3–5 years, ATM programmes, recent offerings and their discount to market) — the equity holder pays for runway in shares.

Phase 5 — Covenants, off-balance-sheet and hybrid instruments

Covenant headroom (only if the credit agreement discloses the tests):

Headroom (%) = (Covenant limit − Actual) / Covenant limit        for a maximum-leverage test
EBITDA cushion = 1 − (Debt / Covenant max leverage) / EBITDA     → how far EBITDA can fall before a breach

Use the covenant's own EBITDA definition (usually more generous than GAAP). Flag headroom < 15%, any amendment or waiver in the last 2 years, and "springing" covenants that test only when the revolver is drawn. If covenants are not disclosed, say "covenants not disclosed — headroom not assessed" (most investment-grade bonds have none).

Debt that is not called debt:

ItemWhereTreatment
Operating leasesLease note (ASC 842)Add the lease liability to adjusted debt (as S&P and Moody's do)
Pension / OPEB deficitPension noteAdd the after-tax underfunded amount (PBO − plan assets)
Guarantees, letters of creditCommitments noteDisclose; add if a call on the guarantee is plausible
Purchase / take-or-pay obligationsContractual obligationsFixed claims on future cash — report next 3 years
Supplier finance / receivables factoringSupplier-finance note (ASU 2022-04), MD&AEconomically debt; flatters OCF and payables — add to adjusted debt and say so
Unconsolidated JV / VIE debtEquity-method investment noteDisclose; add the share the company supports

Convertibles and hybrids. For each convertible: conversion price vs. current price, dilution if converted (shares and % of current count), capped call coverage, and put/maturity date. A convert deep out of the money is plain debt with a near-term maturity; one in the money is equity dilution. Preferred stock and hybrid securities rank ahead of common — include them in the claims ahead of the equity.

Phase 6 — Ratings, spreads and the implied rating

Report the S&P / Moody's / Fitch rating, outlook (positive / stable / negative / watch) and the date of the last action. Report the company's bond spread or CDS if available, and compare it with the median spread for its rating band — a bond trading wide of its rating is the market saying a downgrade is coming.

Rating-band mapping — a rough guide only. The ranges below are approximate, assume an average business-risk profile, and are adapted from the concept of the ratio tables in S&P Global Ratings' corporate methodology and Moody's industry rating methodologies. Actual ratings weight business risk, industry, scale and policy heavily; a strong-business company can carry more leverage at the same rating. Use the table to produce an implied band, never to second-guess an agency to the notch.

BandDebt / EBITDAEBITDA / interestFFO / debtFCF / debtTypical reading for the equity
AAA / AA< 1.0×> 15×> 60%> 40%Debt is irrelevant to the equity case
A1.0 – 2.0×10 – 15×45 – 60%25 – 40%Comfortable
BBB2.0 – 3.0×6 – 10×30 – 45%15 – 25%Investment grade; watch the BBB− edge (a downgrade to high yield forces selling)
BB3.0 – 4.0×3 – 6×20 – 30%8 – 15%Leverage is part of the thesis
B4.0 – 6.0×1.5 – 3×12 – 20%3 – 8%Refinancing access drives the equity
CCC and below> 6.0×< 1.5×< 12%< 3%Equity is an option on survival

When the agency rating and the implied band differ by more than one band, say which way and why (adjusted EBITDA, parent support, a pending deleveraging plan).

Phase 7 — Equity-holder verdict

Translate the credit picture into one of four verdicts:

VerdictConditionsWhat it means for the stock
Residual of a healthy firmImplied band BBB or better, wall coverage > 2× over 3 years, shock < 5% of EBITDebt does not drive the equity case — value the business
Levered but serviceableBB-ish, coverage > 3×, no wall in 2 yearsLeverage magnifies returns both ways — deleveraging is a catalyst, a recession is the risk
Option on refinancingWall coverage < 1× in 2–3 years or shock > 15% of EBIT, but markets are open to the creditThe stock's outcome depends on terms it does not control — price in higher interest and possible dilution
Option on survivalCCC-ish metrics, runway < 6 quarters, covenant breach risk, or bonds trading at distressed levels (spread > 1,000 bp or price < 70)The equity is an out-of-the-money call on the enterprise; in a restructuring it is usually wiped out or heavily diluted. Size it as an option, not an investment

Also note any creditor–equity conflict: liability-management exercises (uptiers, drop-downs of collateral), debt-for-equity exchanges, or a dividend/buyback programme the debt cannot support.

Phase 8 — Credit Health Score (0–10, higher = healthier)

Score each component 0–10 on the anchors, then weight:

ComponentWeight10 looks like0 looks like
Leverage (net debt / EBITDA, adjusted debt)20%Net cash, or < 1.0×> 6.0×, or EBITDA negative
Coverage (EBIT and EBITDA / interest, trend)20%EBIT / interest > 15× and stableEBIT / interest < 1.5× or falling three years running
Maturity wall & liquidity25%3-year wall coverage > 2×, long-dated revolver12-month wall coverage < 0.5× with no committed refinancing
Refinancing cost shock15%Shock < 5% of EBIT (or negative)Shock > 25% of EBIT
Structure & hidden obligations (floating mix, covenants, off-balance-sheet, converts)10%Mostly fixed, unsecured, no tight covenants, small leases/pensionTight covenants (< 15% headroom), large supplier finance or pension deficit, secured debt priming the unsecured
Market signal (rating, outlook, spreads)10%A-range or better, stable, trading tightCCC-range, negative watch, or distressed bond prices

Pre-profit weights: Liquidity runway & dilution 50% · Maturity wall 25% · Structure 15% · Market signal 10% (leverage, coverage and refinancing shock dropped — state it).

Hard caps (state any that fired):

  • Going-concern paragraph, a covenant breach, or a missed payment → Score ≤ 2.0
  • 12-month maturities exceed cash + undrawn revolver + FCF with no committed refinancing, or runway < 4 quarters → Score ≤ 3.0
  • Covenant waiver or amendment to loosen a test in the last 12 months → Score ≤ 5.0

Missing data. Drop any component that cannot be computed, renormalise the remaining weights to 100%, list what was dropped, and cap Confidence at MEDIUM. Never score an unassessed component as 10 or as 0.


3. Output Format

  1. Data & Sources header (above), listing the filings, credit agreement, rating releases and FRED series used, each with its date
  2. Executive Summary — three sentences: the Credit Health Score and verdict, the single biggest credit risk, what would change it
  3. Debt schedule — one row per instrument; fixed vs. floating mix and the +100 bp sensitivity
  4. Maturity wall — 1 / 2 / 3 / 5-year table with coverage ratios
  5. Refinancing shock — tranche-by-tranche ΔInterest, total $, % of EBIT, EPS impact
  6. Ratios — coverage, leverage, FFO / debt, FCF / debt: 5-year trend plus stress case; implied rating band vs. agency rating
  7. Liquidity runway & dilution (pre-profit) or covenant headroom (where disclosed)
  8. Hidden obligations & hybrids — leases, pension, guarantees, supplier finance, converts (dilution if converted)
  9. Ratings & spreads — rating, outlook, last action, spread vs. band median
  10. Equity-holder verdict — one of the four, with the conditions that put it there
  11. Credit Health Score — component table, any cap fired, score
  12. Thesis Invalidation — what would reverse the call (below)
  13. Investment Signal Block — the standard box (below)

Example

User: Credit check on ZEPH (fictional ticker; 10-K debt note and MD&A pasted)

The assistant fills the Data & Sources header (10-K FY2025 pasted by user; BB
effective yield from FRED as of the run date; HIGH for the filing, MEDIUM for the
market yield), then builds the schedule. Hypothetical figures: $1,500M of debt —
a $500M 2.5% note due in 18 months, a $600M 3.75% note due in 4 years, and a $400M
SOFR + 2.25% term loan due in 5 years (27% floating). Cash is $250M, the $300M
revolver expires in 3 years, and base-case FCF is $120M a year.

The 2-year wall coverage is (250 + 300 + 240) / 500 = 1.6× — manageable. Rolling
the 2.5% note at an assumed 7.0% BB yield adds $500M × 4.5% = $22.5M a year of
interest, 9% of EBIT of $250M (≈ $0.13 per share after a 21% tax rate on 140M
shares). EBITDA of $350M gives net debt / EBITDA of 3.6×; interest of about
$61M (term loan at an assumed 6.55% all-in) gives EBITDA / interest of 5.7×:
an implied BB band, matching the agency rating. There are no maintenance
covenants on the notes; the term loan's 4.5× leverage test has 20% headroom.
Operating leases add $180M to adjusted debt.

Components: leverage 5, coverage 6, wall 7, shock 6, structure 6, market 6
→ Credit Health Score 6.1, verdict "levered but serviceable" — BULLISH at the
low end of Moderately Bullish, WEAK-to-MODERATE conviction. Next steps:
`bear-case` with a −30% EBITDA scenario, `dividend-analysis` on whether
the buyback should give way to debt paydown.

Notes

  • Equity view, not a bond recommendation. This skill does not say whether to buy the bonds; it says what the debt means for the stock. A company can be a fine credit and a poor equity (low growth, fully valued) and vice versa.
  • Never invent debt terms. Coupons, maturities and covenant levels come from the filing or the indenture. If the footnote was not retrieved or pasted, mark the field "not verified" — a wrong maturity date changes the whole verdict.
  • Always date the market yields. The refinancing shock is only as current as the yield used; state the series and date, and show the result at ±100 bp.
  • Adjusted EBITDA is the borrower's number. Show it, but score on GAAP-based EBITDA unless the covenant test is the question.
  • Not for banks, insurers or broker-dealers. Their debt is raw material, not financing, and leverage ratios built for operating companies are meaningless. Point to the regulatory capital metrics instead — banks: CET1, Tier 1 leverage, LCR, NSFR, deposit mix, unrealised securities losses; insurers: RBC ratio / Solvency II ratio, reserve adequacy, investment-portfolio credit quality — and use stock-eval --sector bank or --sector insurer. REITs and utilities can be assessed here, but compare with sector norms (utilities carry structurally higher leverage at investment-grade ratings; REITs are judged on net debt / EBITDA and fixed-charge coverage).
  • Cyclicals: use mid-cycle or trough EBITDA in the stress case — peak-cycle leverage always looks fine.
  • Pairs with: financial-report-analyst (the full filing read), stock-eval (whether the business is worth owning), bear-case (the refinancing or recession short thesis), risk-stress-test (the price shock to go with the balance-sheet shock), forensic-accounting (whether the EBITDA itself is reliable), dividend-analysis (whether payouts are crowding out deleveraging).

Thesis Invalidation

After delivering the analysis signal, specify what would reverse it:

If signal is BULLISH (healthy credit) — thesis breaks if:

  • EBITDA falls enough that net debt / EBITDA rises a full turn or EBIT / interest drops below 4× in the next two quarters
  • A debt-funded acquisition, special dividend or buyback raises leverage into the next rating band, or a rating agency moves the outlook to negative
  • The company's bond spread widens by more than 150 bp relative to its rating-band index
  • A maturity within 12 months is not refinanced by six months before it falls due, or the revolver is drawn heavily

If signal is BEARISH (weak credit) — thesis breaks if:

  • The company refinances the near-term wall on terms close to its existing coupons, or extends maturities beyond the horizon
  • An equity raise, asset sale or FCF beat cuts net debt / EBITDA by a full turn
  • Covenant headroom is restored without a waiver, or a rating outlook moves to positive
  • The bond price recovers above 90 / the spread tightens back inside its rating-band median

Re-run this analysis when:

  • Next 10-Q or 10-K (updated maturities, liquidity and covenant compliance)
  • Any 8-K on new debt, a tender/exchange offer, or a credit-agreement amendment
  • A rating action or outlook change by S&P, Moody's or Fitch
  • Price moves ±15% from current level, or rating-band yields move ±100 bp
  • Material news event (acquisition, leadership change, regulatory decision)

Standard Signal Output

This skill measures balance-sheet health from the equity holder's seat, not valuation, so state the mapping: Signal: BULLISH means the debt does not threaten the equity over the horizon; NEUTRAL means leverage is serviceable but a refinancing, rate or EBITDA shock would matter; BEARISH means the equity's outcome depends on creditors or capital markets. Action: SELL here means "the balance sheet does not support holding this equity at an investment-sized position" — read whether the stock is cheap from stock-valuation, not from this block.

All analysis concludes with this standardized block:

In the output, the box below is followed immediately by this machine-readable footer — the same values as the box, filled in (no placeholders, no comments), so tools can read the result without parsing the box:

{"investskill": "1", "skill": "credit-check", "ticker": "<TICKER, or null for a portfolio / market-wide view>", "as_of": "<YYYY-MM-DD from Data & Sources>", "signal": "<BULLISH | NEUTRAL | BEARISH>", "confidence": "<HIGH | MEDIUM | LOW>", "horizon": "<SHORT-TERM | MEDIUM-TERM | LONG-TERM>", "score": 0.0, "action": "<BUY | HOLD | SELL>", "conviction": "<STRONG | MODERATE | WEAK>", "retrieval": "<pasted by user | web/tool retrieval | model memory>", "sources": ["<each primary source, dated>"]}
╔══════════════════════════════════════════════╗
║              INVESTMENT SIGNAL               ║
╠══════════════════════════════════════════════╣
║ Signal:      BULLISH / NEUTRAL / BEARISH     ║
║ Confidence:  HIGH / MEDIUM / LOW             ║
║ Horizon:     SHORT / MEDIUM / LONG-TERM      ║
║ Score:       X.X / 10                        ║
╠══════════════════════════════════════════════╣
║ Action:      BUY / HOLD / SELL               ║
║ Conviction:  STRONG / MODERATE / WEAK        ║
╚══════════════════════════════════════════════╝

Score Guide: 8.0–10.0 Strongly Bullish | 6.0–7.9 Moderately Bullish | 4.0–5.9 Neutral | 2.0–3.9 Moderately Bearish | 0.0–1.9 Strongly Bearish Confidence: HIGH (strong data, clear signals) | MEDIUM (mixed signals) | LOW (limited data, conflicting signals) Horizon: SHORT-TERM (1 week–3 months) | MEDIUM-TERM (3 months–1 year) | LONG-TERM (1+ years)

Note: The box Score is the Credit Health Score itself (higher = healthier balance sheet), so it maps onto the Score Guide directly. Confidence is at most MEDIUM when the debt footnote or market yields were not retrieved or any component was dropped, and LOW when debt terms came from model memory.

Disclaimer: Educational analysis only. Not financial advice.

DCF Valuation

This skill has been merged into stock-valuation. Use /us-stock-analysis:stock-valuation for comprehensive valuation including full DCF modeling, WACC decomposition, 3-scenario sensitivity analysis, and comparable company analysis.


Quick DCF Reference

WACC Formula and Components

WACC = Ke × (E/V) + Kd × (D/V)

Where:
  Ke  = Cost of Equity  = Rf + β × (Rm − Rf) + Size Premium
  Kd  = After-Tax Cost of Debt  = (Interest Expense / Total Debt) × (1 − Tax Rate)
  E/V = Equity Weight  = Equity Market Cap / (Equity Market Cap + Total Debt)
  D/V = Debt Weight    = Total Debt / (Equity Market Cap + Total Debt)

Typical WACC Ranges by Risk Profile:
Risk Profile         WACC Range    Examples
─────────────────────────────────────────────────────
Low risk (utility)   6–8%          Regulated utilities, large cap staples
Medium risk          8–11%         Large cap tech, established growth
High risk            11–15%        Small cap, emerging market, cyclical
Very high risk       15–20%+       Early-stage, distressed, pre-revenue

Three-Scenario Names and Default Probabilities

Scenario    Probability    Narrative
Bull         20%           Market share gains, margin expansion, favorable macro
Base         60%           Historical trend continuation, guidance-aligned
Bear         20%           Competitive pressure, margin compression, macro headwinds

Probability-Weighted IV = (20% × Bull IV) + (60% × Base IV) + (20% × Bear IV)

5×5 Sensitivity Table Template

Sensitivity Table — Intrinsic Value per Share ($)
              Terminal Growth Rate
WACC     1.0%    1.5%    2.0%    2.5%    3.0%
6.0%     $xxx    $xxx    $xxx    $xxx    $xxx
7.0%     $xxx    $xxx    $xxx    $xxx    $xxx
8.0%     $xxx    $xxx    $xxx    $xxx    $xxx  ← Base Case
9.0%     $xxx    $xxx    $xxx    $xxx    $xxx
10.0%    $xxx    $xxx    $xxx    $xxx    $xxx

Terminal Value Formula

Gordon Growth Model:  TV = FCF₁₀ × (1 + g) / (WACC − g)
Exit Multiple Method: TV = FCFₙ × (EV / FCF exit multiple)

Rule: Never set g > WACC. Flag if TV > 80% of total Enterprise Value.

Thesis Invalidation

After delivering the analysis signal, specify what would reverse it:

If signal is BULLISH — thesis breaks if:

  • FCF turns negative for 2 consecutive quarters OR WACC rises >200bps unexpectedly
  • [One or two more triggers drawn from this analysis's own drivers, each with a threshold]

If signal is BEARISH — thesis breaks if:

  • FCF growth accelerates >20% above model assumptions OR interest rates fall >100bps
  • [One or two more triggers drawn from this analysis's own drivers, each with a threshold]

Re-run this analysis when:

  • Next earnings release
  • Price moves ±15% from current level
  • 60 days have elapsed
  • Material news event (acquisition, leadership change, regulatory decision)

Standard Signal Output

All analysis concludes with this standardized block:

In the output, the box below is followed immediately by this machine-readable footer — the same values as the box, filled in (no placeholders, no comments), so tools can read the result without parsing the box:

{"investskill": "1", "skill": "dcf-valuation", "ticker": "<TICKER, or null for a portfolio / market-wide view>", "as_of": "<YYYY-MM-DD from Data & Sources>", "signal": "<BULLISH | NEUTRAL | BEARISH>", "confidence": "<HIGH | MEDIUM | LOW>", "horizon": "<SHORT-TERM | MEDIUM-TERM | LONG-TERM>", "score": 0.0, "action": "<BUY | HOLD | SELL>", "conviction": "<STRONG | MODERATE | WEAK>", "retrieval": "<pasted by user | web/tool retrieval | model memory>", "sources": ["<each primary source, dated>"]}
╔══════════════════════════════════════════════╗
║              INVESTMENT SIGNAL               ║
╠══════════════════════════════════════════════╣
║ Signal:      BULLISH / NEUTRAL / BEARISH     ║
║ Confidence:  HIGH / MEDIUM / LOW             ║
║ Horizon:     SHORT / MEDIUM / LONG-TERM      ║
║ Score:       X.X / 10                        ║
╠══════════════════════════════════════════════╣
║ Action:      BUY / HOLD / SELL               ║
║ Conviction:  STRONG / MODERATE / WEAK        ║
╚══════════════════════════════════════════════╝

Score Guide: 8.0–10.0 Strongly Bullish | 6.0–7.9 Moderately Bullish | 4.0–5.9 Neutral | 2.0–3.9 Moderately Bearish | 0.0–1.9 Strongly Bearish Confidence: HIGH (strong data, clear signals) | MEDIUM (mixed signals) | LOW (limited data, conflicting signals) Horizon: SHORT-TERM (1 week–3 months) | MEDIUM-TERM (3 months–1 year) | LONG-TERM (1+ years)

Disclaimer: Educational analysis only. Not financial advice.

Capital Allocation Analysis (Dividend, Buyback, M&A & FCF Deployment)

⚠️ Data Verification — Do This Before Any Analysis

Before running any analysis, always retrieve the latest market data for the ticker:

  1. Fetch current price — use web search or ask the user for the live price, 52-week range, and market cap. Never assume a price from training data.
  2. Confirm key figures — the declared dividend per share and its ex-date, the last 4 quarters of payments, TTM EPS and FCF (for payout ratios), buyback dollars and share count change, and net debt — from the 10-K/10-Q and the IR dividend history.
  3. State your data source — fill in the Data & Sources header (next section) so the origin, as-of date, retrieval path, and confidence of every figure are explicit at the top of the output.
  4. Flag stale data explicitly — if live data is unavailable, display this warning before proceeding:

⚠️ Live data unavailable. The following analysis uses training-data estimates which may be significantly out of date. Verify all prices and metrics before making any decisions.

Never silently substitute training-data estimates for current prices. When in doubt, ask the user to paste the latest quote.


📋 Data & Sources Header — Open Every Output With It

The first thing in the output is this provenance block, filled in — never left as placeholders. It is the standard documented on the Data & Accuracy page and the first thing result-validator looks for:

Data & Sources
  As of:      <date the figures represent, e.g. 2026-06-30>
  Source:     <primary docs — SEC EDGAR 10-K/10-Q, company IR, FRED, exchange data …>
  Retrieval:  <pasted by user | web/tool retrieval | model memory>
  Confidence: <HIGH | MEDIUM | LOW>
  • Retrieval: model memory must be paired with Confidence: LOW — memory is a placeholder until confirmed against a primary source.
  • Mixed sources: list each with its own as-of date rather than blending them.
  • Sources disagree on the same figure (a closing price, a deal term): prefer the primary source (SEC filing, company release, exchange data) over an aggregator, then the most recent dated one. Show both values with their sources and dates, say which you used and why, and never average them.
  • Data the user pasted is reported as pasted by user; do not upgrade its confidence beyond what the user's own source supports.

🌐 Output Language — --lang zh-TW

English is the default. With --lang zh-TW — or when the user writes in Traditional Chinese and asks for a Chinese answer — write the whole analysis in Traditional Chinese (繁體中文, Taiwan usage), keeping it machine-checkable:

  • Translate every heading, narrative paragraph, table header and bullet. Give each metric's English name in brackets the first time it appears — 自由現金流(Free Cash Flow, FCF) — so the reader can match it to the glossary and to English analyses.
  • Keep in English tickers, figures and units, SEC form and item names (10-K Item 7, Form 4), the four Data & Sources field labels (As of, Source, Retrieval, Confidence — the Retrieval value may be written in Chinese, e.g. 使用者提供), and the Investment Signal box and JSON footer exactly as specified — field names and the BULLISH / NEUTRAL / BEARISH, HIGH / MEDIUM / LOW, BUY / HOLD / SELL, STRONG / MODERATE / WEAK values. Tools parse them.
  • Title the invalidation section 論點失效條件(Thesis Invalidation), and close with the disclaimer in Chinese: 免責聲明: 僅供教育用途,非投資建議。
  • Use Taiwan financial terms: 殖利率、本益比、營業利益率、自由現金流、聯準會、選擇權、買權/賣權、除息日、預扣稅.

Comprehensive capital allocation analysis covering dividend safety, growth trajectory, share buyback discipline, M&A track record, debt management, and FCF deployment quality for US-listed stocks, REITs, and income-focused portfolios.

Analysis Framework

1. Dividend Safety Analysis

Assess the reliability and sustainability of the dividend:

Payout Ratio Analysis

  • EPS-based payout ratio: Dividends per share / Earnings per share. Simple but earnings can be distorted by non-cash items.
  • FCF-based payout ratio: Dividends paid / Free cash flow. More reliable measure — cash must be available to actually pay dividends.
  • AFFO payout ratio (REITs): Dividends / Adjusted Funds From Operations. Standard metric for REITs since depreciation distorts net income.

Safety Thresholds by Sector

Sector               Very Safe    Moderate     Elevated     Danger
General (FCF)          <50%        50-70%       70-85%       >85%
Utilities (FCF)        <60%        60-75%       75-85%       >85%
REITs (AFFO)           <70%        70-80%       80-90%       >90%
Banks (Earnings)       <30%        30-40%       40-55%       >55%
MLPs (DCF)             <60%        60-75%       75-90%       >90%

Dividend Safety Score (0-100)

Weighted composite score:

  • FCF payout ratio (25 pts): <50% = 25, 50-70% = 20, 70-85% = 10, >85% = 0
  • FCF coverage ratio (20 pts): >2x = 20, 1.5-2x = 15, 1.0-1.5x = 8, <1.0x = 0
  • Debt-to-EBITDA (20 pts): <1.5x = 20, 1.5-2.5x = 14, 2.5-3.5x = 7, >3.5x = 0
  • Earnings stability (20 pts): Positive EPS 5yr = 20, 1 down year = 12, 2+ down years = 4
  • Dividend history (15 pts): 10+ yr streak = 15, 5-9 yrs = 10, 2-4 yrs = 5, <2 yrs = 0
Score      Grade    Safety Assessment
90-100       A+     Very Safe
75-89        A      Safe
60-74        B      Borderline Safe
45-59        C      Elevated Risk
30-44        D      Unsafe
0-29         F      Danger Zone

Dividend Stress Test

  • Scenario 1: Earnings decline 20% — can dividend be maintained at current payout?
  • Scenario 2: Earnings decline 40% — what happens to dividend?
  • Scenario 3: FCF drops to 5-year trough — payout ratio at trough FCF?
  • Scenario 4: Revenue declines to 2020 COVID levels — stress test against prior recession data
  • Pass/Fail for each scenario with projected payout ratio under stress

2. Dividend Growth Analysis

Measure the trajectory and sustainability of dividend increases:

Dividend Growth Rate Calculations

  • 1-year DGR: Most recent annual dividend / Prior year annual dividend - 1
  • 3-year DGR CAGR: (Current annual dividend / Dividend 3 years ago)^(1/3) - 1
  • 5-year DGR CAGR: (Current / 5yr ago)^(1/5) - 1
  • 10-year DGR CAGR: (Current / 10yr ago)^(1/10) - 1
DGR Tier        Range          Characteristic
Exceptional      >15%          High-growth compounders (FAST Graphs category)
Strong           8-15%         Solid dividend growers, re-rated higher
Moderate         4-8%          In line with or above inflation
Slow             1-4%          Token increases, inflation parity risk
Frozen           0%            No recent growth
Cut             <0%            Dividend reduced — major red flag

Dividend Aristocrats and Kings

  • Dividend Aristocrats: S&P 500 constituents with 25+ consecutive years of dividend increases
  • Dividend Kings: Stocks with 50+ consecutive years of consecutive increases (elite tier)
  • Dividend Achievers: Stocks with 10+ consecutive years of increases (Nasdaq definition)
  • Status confirmation adds credibility to safety assessment

Chowder Rule

Chowder Number = Current Yield + 5-Year DGR

Thresholds:
- Growth stocks (yield <3%): Chowder Number >= 12%
- High-yield stocks (yield >= 3%): Chowder Number >= 8%
- Utilities: Chowder Number >= 8% (special lower threshold)

Example: Yield 2.5% + 5yr DGR 10% = Chowder 12.5% (PASS for growth stock)

Dividend Growth Sustainability Analysis

  • EPS growth rate vs. dividend growth rate: DGR > EPS growth = unsustainable (payout expansion)
  • Payout ratio trend: expanding payout ratio limits future growth capacity
  • FCF per share growth trend (primary driver of long-term DGR)
  • Analyst consensus EPS growth estimate → maximum sustainable DGR = EPS growth + (payout reduction capacity)
  • Revenue growth required to sustain dividend at current margins

3. Yield Analysis

Evaluate current yield attractiveness in historical and relative context:

Current and Forward Yield

  • Trailing yield: Last 12 months dividends paid / Current price
  • Forward yield: Projected next 12 months dividends / Current price (based on most recent quarterly dividend × 4)
  • Yield spread: Forward yield minus 10-year Treasury yield. Positive spread = attractive income premium vs. risk-free rate.

Historical Yield Context

Yield Position          Interpretation
Current yield < 5yr avg   Stock trading at premium to historical (yield compressed = expensive)
Current yield ≈ 5yr avg   Fairly valued relative to history
Current yield > 5yr avg   Stock trading at discount (yield elevated = potentially cheap or risk elevated)
Current yield > 10yr avg  Historically cheap zone (requires safety check)

Yield vs. 10-Year Treasury Analysis

  • Yield spread over 10-year Treasury: track historical spread compression/expansion
  • Equity risk premium: compensates for equity risk vs. guaranteed government yield
  • When spread < 1%: dividend yield barely compensates for equity risk vs. bonds
  • When spread > 3%: significantly better income from equity vs. bonds (attractive)

Yield-on-Cost (YOC) for Existing Holders

  • YOC = Original purchase price yield × (1 + DGR)^years held
  • Demonstrates power of growing dividends on a fixed cost basis
  • Example: 2% yield at purchase with 10% DGR for 10 years = 5.2% YOC

Yield Trap Detection

High yield + deteriorating business = value trap. Scrutiny triggers:

  • Yield exceeds 7%: require thorough FCF analysis before investing
  • Yield > 2x sector average: market pricing in dividend risk
  • Yield spiked due to price decline (not dividend increase): investigate cause
  • Consecutive quarters of FCF deterioration while yield elevated
  • Debt issuance to fund dividend payments
  • Red flags: declining revenue, rising payout ratio, credit rating downgrades, management tone change on dividend

4. Dividend History and Reliability

Assess the track record of consistent payments:

Payment History Metrics

  • Consecutive years of uninterrupted dividend payments
  • Consecutive years of dividend increases (key Aristocrat/King qualifier)
  • Longest streak before any interruption

Recession Durability

  • 2000-2002 dot-com recession: Was dividend maintained? Cut? Raised?
  • 2008-2009 financial crisis: Stress test benchmark — worst modern recession for dividends
  • 2020 COVID-19 pandemic: Industry-specific stress (travel, retail vs. tech, healthcare)
  • Pattern: Companies that maintained dividends in 2008-2009 AND 2020 = highest quality

Dividend Variability Score

  • Standard deviation of quarterly dividend payments over 5 years
  • Low variability = consistent, predictable income
  • High variability = irregular payments (often MLPs, resource companies)
  • Flag: Any quarters with 0 dividend (interrupted streak)

Special Dividends History

  • Frequency of special/supplemental dividends
  • Size relative to regular dividend (>25% = meaningful supplement)
  • Source: excess FCF, asset sales, one-time items
  • Interpretation: signals strong balance sheet but not guaranteed income

5. Financial Health Supporting Dividends

Analyze the underlying balance sheet and cash flow capacity:

Free Cash Flow Coverage

  • FCF coverage = FCF / Total dividends paid
  • Target: >1.5x (dividend consumes <67% of FCF)
  • Warning: <1.2x (thin margin of safety)
  • Danger: <1.0x (dividend exceeds FCF — funded by debt or asset sales)

Leverage Analysis

Debt-to-EBITDA    Dividend Capacity
<2.0x             Comfortable — dividend growth well supported
2.0-3.0x          Moderate — growth may slow, dividend stable
3.0-4.0x          Constrained — dividend at risk if earnings decline
>4.0x             High risk — debt servicing may crowd out dividends

Interest Coverage Ratio

  • EBIT / Interest expense
  • 5x: Strong — debt servicing leaves ample room for dividends

  • 3-5x: Adequate — moderate buffer
  • 2-3x: Tight — interest burden limits flexibility
  • <2x: Concerning — dividends may compete with debt service

Liquidity Analysis

  • Cash and equivalents on balance sheet
  • Months of dividends covered by current cash (Cash / Annual dividend)
  • Revolving credit facility availability
  • Upcoming debt maturities that require cash allocation

Credit Rating Impact

  • Investment-grade rating (BBB- and above): access to capital markets supports dividend
  • Below investment-grade: higher borrowing cost constrains dividend flexibility
  • Rating watch negative: proactive concern — dividend may be reviewed
  • Downgrade history relative to dividend decisions

6. Income Optimization Analysis

Model the income generation and compounding potential:

Total Return Decomposition

  • Income component: dividend yield contribution to annual return
  • Price appreciation component: capital gains contribution
  • For dividend investors: income component often 30-60% of total return over time

DRIP Reinvestment Analysis

  • Compound growth projection assuming dividends reinvested at current yield
  • Year 1, 5, 10, 20 projections at: 0% DGR (frozen), 3% DGR, 7% DGR, 10% DGR
  • Example: $10,000 investment at 3% yield, 7% DGR, DRIP for 20 years = [calculated value]
  • DRIP accelerates wealth building by purchasing additional shares at each dividend

Tax Efficiency

  • Qualified dividends: taxed at lower capital gains rates (0%, 15%, or 20%)
    • Requirements: US corporation, held >60 days in 120-day window around ex-div
  • Non-qualified (ordinary) dividends: taxed at ordinary income rates (up to 37%)
  • REIT distributions: largely ordinary income (not qualified) — best in tax-advantaged accounts
  • Foreign withholding taxes: may apply to ADRs and foreign-domiciled companies
  • MLP distributions: return of capital treatment (reduces cost basis)

Ex-Dividend Date Calendar

  • Ex-dividend date: must own shares before this date to receive dividend
  • Record date: typically 1 business day after ex-div
  • Payment date: typically 2-4 weeks after record date
  • Impact: stock typically declines by approximately dividend amount on ex-div date

Portfolio Income Modeling

  • At current prices and yields: projected annual income from portfolio
  • Annual income per $100,000 invested by ticker
  • Weighted average portfolio yield
  • Quarterly income distribution timeline

7. Peer Comparison

Benchmark the stock's dividend metrics against sector:

Comparison Metrics Table

Metric[Stock]Sector MedianSector Top QuartileAssessment
Dividend YieldX.X%X.X%X.X%Above/Below
FCF Payout RatioXX%XX%XX%Safe/Risky
5-yr DGRX.X%X.X%X.X%Strong/Weak
Chowder NumberXX.XXX.XXX.XPass/Fail
Safety ScoreXXXXXXGrade
Consecutive IncreasesXX yrsXX yrsXX yrs—

Value vs. Yield Matrix

  • Identify sector peers with better yield at similar or lower valuation
  • Compare P/E vs. yield to identify mispriced dividend payers
  • Best-in-class: highest Chowder Number, highest Safety Score, lowest payout ratio

Capital Allocation — Beyond Dividends

8. Share Buyback Analysis

Evaluate the quality, discipline, and shareholder value impact of the buyback program:

Buyback Authorization vs. Execution Rate

  • Board-authorized repurchase program size ($B and % of market cap)
  • Actual shares repurchased over trailing 1, 3, and 5 years vs. authorization
  • Execution rate = Actual buybacks / Authorized amount. <50% execution signals authorization is more PR than commitment.
  • Open-ended vs. time-limited program structure

Buyback Yield

Buyback Yield = Annual Buybacks ($) / Market Cap

Interpretation:
>5%     Very High — meaningful return of capital
3-5%    High — material shareholder benefit
1-3%    Moderate — supplementary to other returns
<1%     Low — minimal buyback impact

Price Discipline: Are They Buying Smart?

  • Compare average buyback price (total buybacks / shares retired) to estimated intrinsic value
  • Cross-reference buyback timing with stock price history: did they buy at peaks or troughs?
  • Assess whether management references valuation discipline in earnings calls or 8-K filings
  • Red flag: heavy buybacks at peak multiples followed by equity issuance at lower prices (value destruction cycle)
  • Green flag: buybacks accelerate when stock trades below 52-week average and slow at all-time highs

EPS Accretion / Dilution Impact

  • Shares outstanding trend (5-year): shrinking = accretive, flat = offset by stock comp, growing = dilutive
  • Net buyback rate = (Buybacks - Stock-based compensation issuance) / Beginning shares outstanding
  • Positive net buyback rate: genuine per-share value creation
  • Stock-based compensation as % of FCF: >15% indicates compensation is largely offsetting buyback benefits

Insider Ownership Change from Buybacks

  • Management and insider ownership % before and after buyback program
  • Higher ownership % via buybacks (without insider sales) = alignment signal
  • Watch for executives simultaneously selling shares while company repurchases — misalignment flag

9. M&A Capital Allocation

Evaluate how management deploys capital in acquisitions:

Historical Acquisition Multiples Paid

  • List of major acquisitions (last 10 years) with: deal size, EV/EBITDA paid, EV/Revenue paid
  • Compare deal multiples to prevailing sector averages at time of acquisition
  • Premium paid vs. 30-day pre-announcement trading price
Acquisition Multiple Assessment:
EV/EBITDA paid     Assessment
<8x                Disciplined — below sector norm
8-12x              Fair — in line with sector
12-18x             Premium — requires strong strategic rationale
>18x               Rich — significant execution risk, high dilution risk

Acquisition Integration Track Record

  • For each major deal: post-acquisition revenue growth vs. original projections
  • Goodwill impairments taken (a direct admission of overpayment)
  • Post-deal margin trajectory: synergies realized vs. promised synergies?
  • Management tenure on acquired businesses: assets retained or subsequently divested?
  • Rule of thumb: companies that regularly impair goodwill are serial overpayers

Deal Discipline: Overpaying Risk Score

Risk Factor                                    Points
History of goodwill impairments                +2
Average EV/EBITDA paid > sector median + 20%  +2
Acquisitions during peak market periods        +1
Frequent large deals (>3 major in 5 yrs)       +1
Post-deal margin compression                   +1
Management turnover post-acquisition           +1
Overpaying Risk Score: 0 = Disciplined | 3+ = Caution | 5+ = Dealmaker Risk

Organic vs. Inorganic Growth Split

  • Revenue growth decomposed: organic growth % vs. acquisition contribution %
  • Companies growing primarily through acquisitions carry execution and integration risk
  • Preferred profile: >60% organic growth with acquisitions as bolt-ons, not growth substitutes
  • M&A dependency ratio: Acquired revenue in period / Total revenue growth in period

10. Debt Management

Evaluate how management structures and manages the balance sheet:

Debt Paydown Pace vs. Optimal Leverage

  • Current net debt / EBITDA vs. management's stated target leverage
  • Annual debt reduction pace (last 3 years): de-levering or re-levering?
  • Post-acquisition leverage spike: how quickly did they return to target?
  • Optimal leverage range by sector:
Sector              Conservative    Moderate    Stretched
Technology          0-0.5x          0.5-1.5x    >2.0x
Consumer Staples    1.0-2.0x        2.0-3.0x    >3.5x
Industrials         1.5-2.5x        2.5-3.5x    >4.0x
Utilities           2.5-4.0x        4.0-5.0x    >6.0x
REITs               4.0-6.0x        6.0-7.0x    >8.0x

Refinancing Risk (Maturity Schedule)

  • Debt maturity wall: total maturities due in the next 1, 2, 3, and 5 years
  • Maturity concentration: >30% of debt maturing in a single year = elevated refinancing risk
  • Current interest rate environment vs. existing fixed coupon: rising-rate risk on floating debt
  • Undrawn revolving credit facility as buffer against maturity pressure

Covenant Headroom

  • Key financial covenants (Debt/EBITDA, Interest Coverage minimums) from credit agreement disclosures
  • Current ratio vs. covenant threshold: headroom percentage
  • Historical covenant compliance record
  • Waiver history: any covenant waivers obtained = yellow flag

Credit Rating Trend

  • Current rating from Moody's, S&P, Fitch (note most recent action)
  • Rating trajectory (last 3 rating actions: upgrades, downgrades, outlook changes)
  • Investment-grade threshold: BBB-/Baa3 and above — critical for institutional ownership and dividend sustainability
  • Negative outlook or credit watch = potential near-term action risk
  • Spread on bonds vs. comparable investment-grade index: market's implied rating view

11. FCF Deployment Scorecard

Evaluate how every dollar of free cash flow is allocated across competing priorities:

Where Does Every $1 of FCF Go?

Break down actual FCF deployment over trailing 3 years (TTM and 3-year average):

FCF Deployment Breakdown:
  Dividends paid:          XX%   ($X.Xb)
  Share buybacks:          XX%   ($X.Xb)
  Debt reduction:          XX%   ($X.Xb)
  Capital expenditures:    XX%   (already deducted from FCF — note if gross capex used)
  M&A and investments:     XX%   ($X.Xb)
  Cash accumulation:        XX%   ($X.Xb)
  Total:                  100%
  • Note: if using levered FCF, capex is already deducted; use gross cash deployment including capex separately if unlevered FCF is the base
  • Trend: is FCF deployment mix shifting? (e.g., buybacks replacing dividends, or debt paydown replacing buybacks)

Capital Return Yield vs. Peers

Capital Return Yield = Dividend Yield + Buyback Yield

Stock         Div Yield   Buyback Yield   Total Return Yield   vs. Peer Median
[Stock]         X.X%         X.X%              X.X%               +/- X.Xpp
[Peer 1]        X.X%         X.X%              X.X%               Median
[Peer 2]        X.X%         X.X%              X.X%               +/- X.Xpp
[Peer 3]        X.X%         X.X%              X.X%               +/- X.Xpp

Management Capital Allocation Grade

Score each dimension and assign an overall letter grade:

DimensionScoreGradeKey Evidence
Dividend safety & growthX/10A-FPayout ratio, streak, DGR
Buyback disciplineX/10A-FPrice timing, net share reduction
M&A track recordX/10A-FGoodwill impairments, synergy delivery
Debt managementX/10A-FLeverage trajectory, maturity management
FCF deployment efficiencyX/10A-FReturn yield vs. peers, cash hoarding
Overall GradeX/10A-FComposite assessment
Grade Criteria:
A (9-10): Consistent compounders — buyback below fair value, dividend aristocrat, M&A creates value, optimal leverage
B (7-8):  Good stewards — solid on most dimensions, one area of weakness
C (5-6):  Average — market-rate capital return, limited M&A track record
D (3-4):  Poor — overpays for M&A, buybacks at peak, dividend growth stagnant
F (0-2):  Value destroyers — goodwill impairments, dividend cuts, re-levering balance sheet

12. Capital Allocation Quality Score (Composite 0–10)

Single composite score summarizing overall capital allocation quality:

Component                        Weight   Score (0-10)   Weighted Score
Dividend Safety Score              20%      X.X            X.X
Dividend Growth Quality            10%      X.X            X.X
Buyback Discipline                 20%      X.X            X.X
M&A Track Record                   20%      X.X            X.X
Debt Management Quality            15%      X.X            X.X
FCF Deployment Efficiency          15%      X.X            X.X
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
Capital Allocation Quality Score   100%                    X.X / 10
Score    Interpretation
9-10     Exceptional stewardship — rare, compounding machines
7-8      Strong allocators — above-average long-term value creation
5-6      Average — market-rate capital allocation, no clear edge
3-4      Below average — capital misallocation risk weighs on returns
0-2      Poor stewardship — history of value destruction through M&A, buybacks at peaks, or dividend unsustainability

Data Sources

Primary Financials

  • SEC filings (10-K / 10-Q): FCF, net income, dividends paid (cash flow statement), share repurchase disclosures
  • Company investor relations pages: dividend history, buyback program announcements, earnings call transcripts

Dividend Research Platforms

  • Simply Safe Dividends: Dividend Safety Scores, comprehensive dividend data (premium)
  • Dividend.com: Dividend yield, history, DRIP calculators
  • DRIP Investing Resource Center (dripinvesting.org): Aristocrats/Kings lists
  • Seeking Alpha Dividends: Analyst commentary on dividend sustainability
  • S&P Global: Official Dividend Aristocrats and Kings lists

Capital Allocation Research

  • Compustat / FactSet: Multi-year buyback and M&A data
  • Bloomberg: Credit ratings, bond spreads, covenant disclosures
  • Morningstar: Capital allocation ratings, M&A track record, stewardship grade
  • GuruFocus: FCF analysis, buyback history, insider ownership

Screening Tools

  • Finviz screener: Filter by dividend yield, payout ratio, buyback yield
  • Morningstar: Capital Allocation rating (Exemplary / Standard / Poor)
  • GuruFocus: Capital allocation and management quality scores

Input Formats

Format 1: Single Ticker

User: /dividend-analysis AAPL

Claude analyzes AAPL across all capital allocation dimensions

Format 2: Portfolio Income Review

User: /dividend-analysis --portfolio AAPL,JNJ,KO,PEP,XOM

Claude provides comparative analysis across all tickers with portfolio-level income model and capital return yields

Format 3: Sector Screen

User: /dividend-analysis --sector utilities

Claude screens utilities sector for best capital allocation quality using safety, growth, buyback yield, and M&A discipline criteria

Output

Provide a comprehensive capital allocation analysis report with the following sections:

1. Executive Summary

  • Capital Allocation Quality Score: [0-10] with interpretation
  • Dividend Safety Score: [0-100] with letter grade (A+/A/B/C/D/F)
  • Safety Assessment: Very Safe / Safe / Borderline / Unsafe / Danger
  • Current Yield: X.X% trailing | X.X% forward
  • Capital Return Yield: X.X% (dividend + buyback yield)
  • Chowder Number: X.X (Pass/Fail)
  • Management Allocation Grade: A/B/C/D/F
  • Key Finding: 2-3 sentence summary of most important conclusion

2. Dividend Safety Analysis

FCF Payout Ratio:       XX%     (Target <70%)
FCF Coverage Ratio:     X.Xx    (Target >1.5x)
Debt-to-EBITDA:         X.Xx    (Target <3x)
Interest Coverage:      X.Xx    (Target >3x)
Stress Test (20% EPS):  PASS/FAIL
Stress Test (40% EPS):  PASS/FAIL
Safety Score:           XX/100  Grade [X]

3. Dividend Growth Metrics

1-Year DGR:             X.X%
3-Year DGR (CAGR):      X.X%
5-Year DGR (CAGR):      X.X%
10-Year DGR (CAGR):     X.X%
Aristocrat Status:      Yes/No (XX consecutive years)
Chowder Number:         X.X% (Pass/Fail at X% threshold)
Payout Ratio Trend:     Expanding / Stable / Contracting

4. Share Buyback Scorecard

Buyback Yield (TTM):    X.X%
Net Share Reduction:    X.X% annualized (shares outstanding trend)
Stock Comp Offset:      XX% of buybacks offset by SBC
Price Discipline:       Buying below / at / above estimated intrinsic value
EPS Accretion Impact:   +X.X% annual EPS lift from net share reduction
Buyback Grade:          A / B / C / D / F

5. M&A Track Record

  • Summary table of major acquisitions with multiples paid and outcome
  • Goodwill impairment history
  • Organic vs. inorganic revenue growth split
  • Overpaying Risk Score: X/8 (Low / Moderate / High)
  • M&A Grade: A / B / C / D / F

6. Debt Management Assessment

Net Debt / EBITDA:      X.Xx  (Target range: X.X-X.Xx)
Management Target:      X.Xx  (stated in investor materials)
Leverage Trend:         De-levering / Stable / Re-levering
Nearest Maturity Wall:  $Xb due in [Year] (XX% of total debt)
Credit Rating:          [Moody's] / [S&P] / Outlook
Covenant Headroom:      XX% above nearest covenant threshold
Debt Grade:             A / B / C / D / F

7. FCF Deployment Breakdown

  • Pie breakdown: dividends %, buybacks %, M&A %, debt paydown %, cash accumulation %
  • Capital return yield vs. 3-5 sector peers
  • Management Capital Allocation Grade table (all 5 dimensions)

8. Capital Allocation Quality Score Summary

Full scoring table with component weights and final 0-10 composite score

9. Yield Trap Assessment

  • Is current yield elevated vs. historical? If yes, why?
  • FCF trend supporting or undermining yield?
  • Red flag checklist (5-7 criteria, checked or clear)
  • Verdict: Genuine Value / Yield Trap Risk / Monitoring Required

10. Peer Comparison Table

Full comparison table vs. 3-5 sector peers across key metrics (dividend yield, buyback yield, total return yield, FCF payout, DGR, capital allocation score)

11. Income Projections

$10,000 Invested at Current Price:
  Annual income (Year 1):    $XXX
  Annual income (Year 5):    $XXX  (at X% projected DGR)
  Annual income (Year 10):   $XXX
  Yield-on-Cost (Year 10):   X.X%
  DRIP value (Year 10):      $XX,XXX

12. Key Risks to Capital Allocation

  • Ranked list of top 3-5 risks (dividend cut, M&A misstep, leverage spike, buyback cessation)
  • Probability assessment (Low/Medium/High) for each risk

13. Monitoring Triggers

  • Specific metrics and thresholds that would change the safety assessment or overall score
  • Next dividend declaration date / ex-div date
  • Upcoming earnings where FCF and buyback data will be updated
  • Key M&A watch: is management signaling appetite for large deals?

Thesis Invalidation

After delivering the analysis signal, specify what would reverse it:

If signal is BULLISH — thesis breaks if:

  • dividend cut announced OR debt/EBITDA exceeds 4x
  • [One or two more triggers drawn from this analysis's own drivers, each with a threshold]

If signal is BEARISH — thesis breaks if:

  • dividend raised AND FCF payout ratio improves below 50%
  • [One or two more triggers drawn from this analysis's own drivers, each with a threshold]

Re-run this analysis when:

  • Next earnings release
  • Price moves ±15% from current level
  • 60 days have elapsed
  • Material news event (acquisition, leadership change, regulatory decision)

Standard Signal Output

All analysis concludes with this standardized block:

In the output, the box below is followed immediately by this machine-readable footer — the same values as the box, filled in (no placeholders, no comments), so tools can read the result without parsing the box:

{"investskill": "1", "skill": "dividend-analysis", "ticker": "<TICKER, or null for a portfolio / market-wide view>", "as_of": "<YYYY-MM-DD from Data & Sources>", "signal": "<BULLISH | NEUTRAL | BEARISH>", "confidence": "<HIGH | MEDIUM | LOW>", "horizon": "<SHORT-TERM | MEDIUM-TERM | LONG-TERM>", "score": 0.0, "action": "<BUY | HOLD | SELL>", "conviction": "<STRONG | MODERATE | WEAK>", "retrieval": "<pasted by user | web/tool retrieval | model memory>", "sources": ["<each primary source, dated>"]}
╔══════════════════════════════════════════════╗
║              INVESTMENT SIGNAL               ║
╠══════════════════════════════════════════════╣
║ Signal:      BULLISH / NEUTRAL / BEARISH     ║
║ Confidence:  HIGH / MEDIUM / LOW             ║
║ Horizon:     SHORT / MEDIUM / LONG-TERM      ║
║ Score:       X.X / 10                        ║
╠══════════════════════════════════════════════╣
║ Action:      BUY / HOLD / SELL               ║
║ Conviction:  STRONG / MODERATE / WEAK        ║
╚══════════════════════════════════════════════╝

Score Guide: 8.0–10.0 Strongly Bullish | 6.0–7.9 Moderately Bullish | 4.0–5.9 Neutral | 2.0–3.9 Moderately Bearish | 0.0–1.9 Strongly Bearish Confidence: HIGH (strong data, clear signals) | MEDIUM (mixed signals) | LOW (limited data, conflicting signals) Horizon: SHORT-TERM (1 week–3 months) | MEDIUM-TERM (3 months–1 year) | LONG-TERM (1+ years)

Disclaimer: Educational analysis only. Not financial advice.

Earnings Call Analysis

⚠️ Data Verification — Do This Before Any Analysis

Before running any analysis, always retrieve the latest market data for the ticker:

  1. Fetch current price — use web search or ask the user for the live price, 52-week range, and market cap. Never assume a price from training data.
  2. Confirm key figures — the transcript's date and quarter, the reported revenue/EPS vs. consensus for that quarter, and the prior quarter's guidance — so tone is judged against what management actually delivered.
  3. State your data source — fill in the Data & Sources header (next section) so the origin, as-of date, retrieval path, and confidence of every figure are explicit at the top of the output.
  4. Flag stale data explicitly — if live data is unavailable, display this warning before proceeding:

⚠️ Live data unavailable. The following analysis uses training-data estimates which may be significantly out of date. Verify all prices and metrics before making any decisions.

Never silently substitute training-data estimates for current prices. When in doubt, ask the user to paste the latest quote.


📋 Data & Sources Header — Open Every Output With It

The first thing in the output is this provenance block, filled in — never left as placeholders. It is the standard documented on the Data & Accuracy page and the first thing result-validator looks for:

Data & Sources
  As of:      <date the figures represent, e.g. 2026-06-30>
  Source:     <primary docs — SEC EDGAR 10-K/10-Q, company IR, FRED, exchange data …>
  Retrieval:  <pasted by user | web/tool retrieval | model memory>
  Confidence: <HIGH | MEDIUM | LOW>
  • Retrieval: model memory must be paired with Confidence: LOW — memory is a placeholder until confirmed against a primary source.
  • Mixed sources: list each with its own as-of date rather than blending them.
  • Sources disagree on the same figure (a closing price, a deal term): prefer the primary source (SEC filing, company release, exchange data) over an aggregator, then the most recent dated one. Show both values with their sources and dates, say which you used and why, and never average them.
  • Data the user pasted is reported as pasted by user; do not upgrade its confidence beyond what the user's own source supports.

🌐 Output Language — --lang zh-TW

English is the default. With --lang zh-TW — or when the user writes in Traditional Chinese and asks for a Chinese answer — write the whole analysis in Traditional Chinese (繁體中文, Taiwan usage), keeping it machine-checkable:

  • Translate every heading, narrative paragraph, table header and bullet. Give each metric's English name in brackets the first time it appears — 自由現金流(Free Cash Flow, FCF) — so the reader can match it to the glossary and to English analyses.
  • Keep in English tickers, figures and units, SEC form and item names (10-K Item 7, Form 4), the four Data & Sources field labels (As of, Source, Retrieval, Confidence — the Retrieval value may be written in Chinese, e.g. 使用者提供), and the Investment Signal box and JSON footer exactly as specified — field names and the BULLISH / NEUTRAL / BEARISH, HIGH / MEDIUM / LOW, BUY / HOLD / SELL, STRONG / MODERATE / WEAK values. Tools parse them.
  • Title the invalidation section 論點失效條件(Thesis Invalidation), and close with the disclaimer in Chinese: 免責聲明: 僅供教育用途,非投資建議。
  • Use Taiwan financial terms: 殖利率、本益比、營業利益率、自由現金流、聯準會、選擇權、買權/賣權、除息日、預扣稅.

Comprehensive analysis of earnings call transcripts to extract investment insights, management sentiment, strategic themes, and potential red flags.

Analysis Framework

1. Executive Summary Analysis

Extract and synthesize:

  • Overall Sentiment: Bullish / Neutral / Bearish
  • Key Takeaways: 3-5 most important points from the call
  • Guidance Changes: Raised / Maintained / Lowered / Withdrawn
  • Surprise Factors: Unexpected announcements or revelations
  • Market-Moving Statements: Comments likely to impact stock price

2. Management Tone Assessment

Analyze management's communication style and confidence:

Confidence Indicators

  • Use of definitive language ("will", "committed", "confident")
  • Specific quantitative guidance with narrow ranges
  • Proactive discussion of challenges with clear solutions
  • Long-term vision and strategic clarity

Caution Indicators

  • Hedging language ("may", "could", "hope", "expect")
  • Wide guidance ranges or lack of specific targets
  • Deflection of difficult questions
  • Focus on short-term rather than strategic issues

Red Flag Indicators

  • Evasive or defensive responses
  • Frequent changes in financial definitions or metrics
  • Blame shifting to external factors
  • Reduced transparency or detail vs. prior quarters

3. Key Themes Extraction

Identify and categorize main topics discussed:

Strategic Initiatives

  • New product launches or market expansion
  • M&A activity and integration plans
  • Digital transformation initiatives
  • Cost reduction or efficiency programs
  • R&D investments and innovation pipeline

Operational Performance

  • Revenue drivers and growth catalysts
  • Margin trends and cost pressures
  • Supply chain and operational challenges
  • Customer trends and demand patterns
  • Geographic or segment performance variations

Market Dynamics

  • Competitive landscape changes
  • Industry trends and macro headwinds/tailwinds
  • Regulatory or policy impacts
  • Technology disruptions or opportunities

Capital Allocation

  • Dividend policy and changes
  • Share buyback plans
  • Capex priorities and timing
  • Debt management and balance sheet strategy

4. Financial Guidance Analysis

Compare and analyze forward-looking statements:

Guidance Components

  • Revenue guidance: Range, growth rate, key assumptions
  • Earnings guidance: EPS range, margin expectations
  • Segment-specific guidance
  • Full-year vs. next-quarter outlook

Guidance Changes from Prior Quarter

  • Magnitude of change (%)
  • Reasons cited for changes
  • Credibility assessment based on track record
  • Comparison to analyst consensus expectations

Key Assumptions

  • Macro assumptions (GDP, rates, FX)
  • Volume and pricing assumptions
  • Cost inflation expectations
  • One-time items or adjustments

5. Q&A Session Analysis

Evaluate analyst questions and management responses:

Question Themes (by frequency)

  1. [Most frequently asked topics]
  2. [Second most common questions]
  3. [Emerging concerns]

Management Response Quality

  • Direct and substantive vs. evasive
  • Quantitative detail provided
  • Consistency with prepared remarks
  • New information revealed vs. scripted responses

Analyst Concerns

  • Recurring questions (indicating lack of clarity)
  • Challenging questions about weaknesses
  • Questions about competitive threats
  • Governance or strategic direction inquiries

Notable Exchanges

  • Tense or defensive moments
  • Particularly illuminating answers
  • Questions management avoided or deflected

6. Quarter-over-Quarter Comparison

Compare with previous earnings calls (if available):

Messaging Consistency

  • Changed narratives or priorities
  • Previously emphasized themes now downplayed
  • New strategic focuses or pivots

Tone Shifts

  • More/less confident than prior quarters
  • Defensive vs. offensive positioning
  • Transparency level changes

Performance vs. Prior Guidance

  • Did they deliver on previous commitments?
  • Track record of guidance accuracy
  • Patterns of over-promising or sandbagging

7. Red Flags & Warning Signs

Identify potential concerns:

Financial Red Flags

  • Missed guidance for consecutive quarters
  • Declining margins despite revenue growth
  • Working capital deterioration
  • FCF significantly below net income
  • One-time charges becoming recurring

Operational Red Flags

  • Customer concentration risks
  • Key executive departures
  • Product delays or quality issues
  • Market share losses
  • Failed M&A integrations

Communication Red Flags

  • Lack of transparency or detail reduction
  • Frequent changes in reporting metrics
  • Aggressive accounting interpretations
  • Dismissive attitude toward legitimate concerns
  • Unrealistic optimism despite evidence

Governance Red Flags

  • Board or management conflicts
  • Related-party transactions
  • Insider selling patterns
  • Lack of succession planning

8. Investment Implications

Synthesize analysis into actionable insights:

Overall Assessment

  • Bullish / Neutral / Bearish sentiment justified by evidence
  • Confidence level in management's strategy and execution
  • Visibility and predictability of future performance

Key Investment Drivers

  • Most important positive factors
  • Most significant risks and concerns
  • Catalysts for outperformance
  • Potential downside scenarios

Valuation Impact

  • Likely direction of analyst estimate revisions
  • Impact on earnings growth rate expectations
  • Multiple expansion/contraction considerations
  • Fair value reassessment based on new information

Recommended Action

  • Buy / Hold / Sell recommendation with rationale
  • Price target adjustment (if applicable)
  • Position sizing considerations
  • Time horizon for thesis to play out

Data Sources

Reference the following sources when analyzing earnings calls:

Primary Sources

  • SEC EDGAR 8-K Filings: Official transcripts filed within 4 days of call

  • Company Investor Relations: Live webcasts and archived recordings

    • Typically at: [company-website.com]/investors/earnings
    • May include slides and supplemental materials

Transcript Services

  • Seeking Alpha: Free transcript archive with searchable database

  • AlphaSense: Premium transcript service with AI search (paid)

  • Bloomberg Terminal: Real-time transcripts and historical archive (paid)

  • FactSet: Transcript archive with analytics (paid)

Supplementary Information

  • Earnings Press Release: Published before call, contains prepared financial results
  • Investor Presentation Slides: Visual aids referenced during call
  • 10-Q/10-K Filings: Detailed financials for context
  • Prior Quarter Calls: For comparison and trend analysis

Input Formats

Accept earnings call data in the following formats:

Format 1: Transcript Text

Paste the full transcript text from the earnings call, including:
- Corporate Participants section
- Prepared Remarks
- Q&A Session
- Forward-Looking Statements disclaimer (if present)

Format 2: Transcript URL

Provide URL to transcript:
- Seeking Alpha URL
- Company IR page URL
- SEC EDGAR filing link

Format 3: Call Details for Manual Analysis

Company: [Name]
Ticker: [SYMBOL]
Quarter: Q[1-4] [YEAR]
Call Date: [MM/DD/YYYY]

Key Questions to Focus On:
- [Specific aspects to analyze]

Output

Provide comprehensive earnings call analysis report with:

1. Call Metadata

  • Company name and ticker
  • Quarter and fiscal year
  • Call date and time
  • Participating executives
  • Transcript source

2. Executive Summary (1-2 paragraphs)

  • Overall assessment and sentiment
  • Key takeaways (3-5 bullet points)
  • Investment implication (Buy/Hold/Sell)

3. Management Tone Rating

  • Confidence Level: High / Medium / Low
  • Transparency: Excellent / Good / Fair / Poor
  • Notable tone characteristics
  • Supporting evidence (quotes)

4. Key Themes (Ranked by Importance)

For each theme:

  • Description
  • Management's stance
  • Supporting quotes
  • Investment relevance

5. Guidance Analysis

  • Detailed guidance breakdown
  • Changes from prior guidance
  • Comparison to consensus estimates
  • Credibility assessment

6. Q&A Insights

  • Top 5 analyst questions
  • Quality of management responses
  • New information revealed
  • Questions avoided or deflected

7. Red Flags & Concerns

  • Identified warning signs (if any)
  • Severity assessment
  • Historical context
  • Recommended monitoring items

8. Quarter-over-Quarter Changes

  • Messaging shifts
  • Tone differences
  • Strategic pivots
  • Performance vs. prior commitments

9. Investment Recommendation

  • Overall sentiment: Bullish / Neutral / Bearish
  • Confidence level: High / Medium / Low
  • Key investment drivers (bullish case)
  • Key risks (bearish case)
  • Suggested action with rationale
  • Time horizon

10. Notable Quotes

5-10 most significant quotes from management with context

Analysis Guidelines

Best Practices

  1. Read the Entire Transcript: Don't rely only on prepared remarks; Q&A often reveals more
  2. Compare to Prior Calls: Context is critical for identifying changes
  3. Cross-Reference with Financials: Verify claims against actual 10-Q/10-K data
  4. Watch for Language Patterns: Hedging, definitiveness, evasiveness
  5. Note What's NOT Said: Absence of previously discussed topics is meaningful
  6. Consider the Audience: Different messages for different analyst types

Common Pitfalls to Avoid

  1. Over-Optimism Bias: Management is inherently promotional
  2. Recency Bias: One quarter doesn't make a trend
  3. Cherry-Picking: Consider full context, not just favorable quotes
  4. Ignoring Tone: What's said matters, but HOW it's said matters more
  5. Missing Red Flags: Defensiveness and evasion are important signals

Analyst Question Interpretation

  • If analysts ask about guidance: They may expect a beat or miss
  • If questions focus on one issue: It's likely a consensus concern
  • If analysts challenge management: Credibility may be in question
  • If questions are softball: Either results are stellar or relationships are managed

Usage Examples

Example 1: Basic Transcript Analysis

User: /earnings-call-analysis AAPL

[Paste transcript text or provide URL]

Claude analyzes the transcript and provides comprehensive report with all sections

Example 2: Focused Analysis

User: /earnings-call-analysis TSLA --focus "production guidance,margin outlook"

[Provide transcript]

Claude emphasizes analysis of specified topics while still covering full call

Example 3: Comparative Analysis

User: /earnings-call-analysis NVDA --compare Q3-2024

[Provide current quarter transcript]

Claude compares with prior quarter (Q3 2024) to highlight changes and trends

Integration Notes

  • Can be used standalone or in conjunction with /stock-eval
  • Output can be fed to /report-generator for HTML/PDF export
  • Particularly valuable around earnings season (Jan, Apr, Jul, Oct)
  • Best used within 1-2 days of earnings release while market is digesting information

Thesis Invalidation

After delivering the analysis signal, specify what would reverse it:

If signal is BULLISH — thesis breaks if:

  • management lowers full-year guidance by >10% OR CEO/CFO departure announced within 30 days
  • [One or two more triggers drawn from this analysis's own drivers, each with a threshold]

If signal is BEARISH — thesis breaks if:

  • major earnings beat >15% AND multi-year guidance raised significantly
  • [One or two more triggers drawn from this analysis's own drivers, each with a threshold]

Re-run this analysis when:

  • Next earnings release
  • Price moves ±15% from current level
  • 60 days have elapsed
  • Material news event (acquisition, leadership change, regulatory decision)

Standard Signal Output

All analysis concludes with this standardized block:

In the output, the box below is followed immediately by this machine-readable footer — the same values as the box, filled in (no placeholders, no comments), so tools can read the result without parsing the box:

{"investskill": "1", "skill": "earnings-call-analysis", "ticker": "<TICKER, or null for a portfolio / market-wide view>", "as_of": "<YYYY-MM-DD from Data & Sources>", "signal": "<BULLISH | NEUTRAL | BEARISH>", "confidence": "<HIGH | MEDIUM | LOW>", "horizon": "<SHORT-TERM | MEDIUM-TERM | LONG-TERM>", "score": 0.0, "action": "<BUY | HOLD | SELL>", "conviction": "<STRONG | MODERATE | WEAK>", "retrieval": "<pasted by user | web/tool retrieval | model memory>", "sources": ["<each primary source, dated>"]}
╔══════════════════════════════════════════════╗
║              INVESTMENT SIGNAL               ║
╠══════════════════════════════════════════════╣
║ Signal:      BULLISH / NEUTRAL / BEARISH     ║
║ Confidence:  HIGH / MEDIUM / LOW             ║
║ Horizon:     SHORT / MEDIUM / LONG-TERM      ║
║ Score:       X.X / 10                        ║
╠══════════════════════════════════════════════╣
║ Action:      BUY / HOLD / SELL               ║
║ Conviction:  STRONG / MODERATE / WEAK        ║
╚══════════════════════════════════════════════╝

Score Guide: 8.0–10.0 Strongly Bullish | 6.0–7.9 Moderately Bullish | 4.0–5.9 Neutral | 2.0–3.9 Moderately Bearish | 0.0–1.9 Strongly Bearish Confidence: HIGH (strong data, clear signals) | MEDIUM (mixed signals) | LOW (limited data, conflicting signals) Horizon: SHORT-TERM (1 week–3 months) | MEDIUM-TERM (3 months–1 year) | LONG-TERM (1+ years)

Disclaimer: Educational analysis only. Not financial advice.

Earnings Preview — What Is Priced In Before the Print

⚠️ Data Verification — Do This Before Any Analysis

Before running any analysis, always retrieve the latest market data for the ticker:

  1. Fetch current price — use web search or ask the user for the live price, 52-week range, and market cap. Never assume a price from training data.
  2. Confirm the event and the expectations — the confirmed earnings date and time (BMO / AMC) from the company's IR page (not a calendar estimate), the current consensus revenue / EPS / guidance with its as-of date, the prior guidance range management gave last quarter, and current option prices for the expiration that spans the print (at-the-money straddle, or implied volatility). Never carry consensus from memory — it moves in the final two weeks.
  3. State your data source — fill in the Data & Sources header (next section) so the origin, as-of date, retrieval path, and confidence of every figure are explicit at the top of the output.
  4. Flag stale data explicitly — if live data is unavailable, display this warning before proceeding:

⚠️ Live data unavailable. The following analysis uses training-data estimates which may be significantly out of date. Verify all prices and metrics before making any decisions.

Never silently substitute training-data estimates for current prices. When in doubt, ask the user to paste the latest quote.


📋 Data & Sources Header — Open Every Output With It

The first thing in the output is this provenance block, filled in — never left as placeholders. It is the standard documented on the Data & Accuracy page and the first thing result-validator looks for:

Data & Sources
  As of:      <date the figures represent, e.g. 2026-06-30>
  Source:     <primary docs — SEC EDGAR 10-K/10-Q, company IR, FRED, exchange data …>
  Retrieval:  <pasted by user | web/tool retrieval | model memory>
  Confidence: <HIGH | MEDIUM | LOW>
  • Retrieval: model memory must be paired with Confidence: LOW — memory is a placeholder until confirmed against a primary source.
  • Mixed sources: list each with its own as-of date rather than blending them.
  • Sources disagree on the same figure (a closing price, a deal term): prefer the primary source (SEC filing, company release, exchange data) over an aggregator, then the most recent dated one. Show both values with their sources and dates, say which you used and why, and never average them.
  • Data the user pasted is reported as pasted by user; do not upgrade its confidence beyond what the user's own source supports.

🌐 Output Language — --lang zh-TW

English is the default. With --lang zh-TW — or when the user writes in Traditional Chinese and asks for a Chinese answer — write the whole analysis in Traditional Chinese (繁體中文, Taiwan usage), keeping it machine-checkable:

  • Translate every heading, narrative paragraph, table header and bullet. Give each metric's English name in brackets the first time it appears — 自由現金流(Free Cash Flow, FCF) — so the reader can match it to the glossary and to English analyses.
  • Keep in English tickers, figures and units, SEC form and item names (10-K Item 7, Form 4), the four Data & Sources field labels (As of, Source, Retrieval, Confidence — the Retrieval value may be written in Chinese, e.g. 使用者提供), and the Investment Signal box and JSON footer exactly as specified — field names and the BULLISH / NEUTRAL / BEARISH, HIGH / MEDIUM / LOW, BUY / HOLD / SELL, STRONG / MODERATE / WEAK values. Tools parse them.
  • Title the invalidation section 論點失效條件(Thesis Invalidation), and close with the disclaimer in Chinese: 免責聲明: 僅供教育用途,非投資建議。
  • Use Taiwan financial terms: 殖利率、本益比、營業利益率、自由現金流、聯準會、選擇權、買權/賣權、除息日、預扣稅.

Overview

The catalog already covers earnings after the fact: earnings-call-analysis reads the transcript, financial-report-analyst reads the 10-Q, catalyst-calendar puts the date on a list. None of them answers the question a holder actually has in the week before the print: what does the market already expect, how has this stock treated good news before, and what am I going to do in each outcome — decided now, while I am calm?

This skill is that pre-commitment. It produces four things no post-call skill can:

  1. The expectation stack — published consensus, the buy-side "whisper" above it, and management's own prior guidance, so a "beat" can be judged against the number that actually matters.
  2. The stock's own history with beats — eight quarters of beat/miss against the next-day move, which is where "beat and drop" patterns live.
  3. What the price already assumes — the growth and margin the current quote requires, so the reader knows whether a beat is news or merely confirmation.
  4. A three-scenario grid with a rule attached to each cell — so the reaction on print day is execution, not decision.

What it is not. It does not predict the print. Nobody knows the number; the skill's value is in knowing what others expect and what you will do. It is also not an options strategy tool — it reads the options market for its implied move and hands strategy selection to options-analysis.

Pairs with earnings-call-analysis (run it after, on the transcript), options-analysis (strategy for the implied move), catalyst-calendar (the date and the events around it), thesis-tracker (the print is a scheduled next-check — the KPIs here should be the KPIs there), and position-ladder (the position rules in the scenario grid should respect its share-count ceiling).


1. Inputs

InputRequiredDefault if unstated
Ticker✅—
Confirmed earnings date and time (BMO / AMC)✅fetch from company IR; if only a calendar estimate exists, say so and mark Confidence MEDIUM at best
Current consensus: revenue, EPS, next-quarter / FY guidance✅fetch or ask; state the as-of date
Management's prior guidance range✅from the last press release / call
Current price, ATM straddle price or IV for the expiry spanning the print✅fetch or ask
Last 8 quarters: reported vs. consensus, next-day moverecommendedfetch or ask; fewer than 6 quarters → note the thin sample
Prior-quarter transcript or earnings-call-analysis outputoptionalimproves the KPI list and the "what management promised" line
Existing thesis-tracker file or stock-eval outputoptionalsupplies the KPIs and thresholds that should decide the reaction
Position: shares, cost basis, ceiling from position-ladderoptionalneeded for the position rules; otherwise rules are stated generically

Whisper numbers are estimates, and are labeled as such. The "whisper" is the buy-side expectation sitting above (rarely below) published consensus — the number the stock actually trades against. It is inferred from sell-side note revisions, the direction of estimate changes in the last 30 days, options skew, and sentiment sources; it is never a hard figure. Report it as a range with a source, or write whisper: not estimable rather than inventing one.


2. Framework

Phase 1 — The expectation stack

Build the table of numbers the print will be judged against, in the order the market judges them:

LayerRevenueEPSNext-Q guideFY guideSource · as of
Management's prior guidance (range and midpoint)
Published consensus
Consensus 30 / 90 days ago (direction of revisions)
Whisper (estimate, range)
Gap: whisper − consensus

Rules of thumb:

  • A rising consensus into the print raises the bar: a beat against a number that has already been marked up is a smaller beat than it looks.
  • Whisper gap > ~2% of revenue or > ~5% of EPS means a consensus-level beat is likely to be treated as a miss. Say so plainly.
  • The guide is the number that moves the stock for most growth companies. If prior guidance exists, the market is asking "raise, hold, or cut?" — not "beat or miss?".

Phase 2 — How this stock treats beats (8-quarter history)

QuarterRev vs. consensusEPS vs. consensusGuide vs. consensusNext-day moveMove vs. implied
Q-1+x.x%+x.x%raised / held / cut±x.x%above / within / below
…

Then summarize:

  • Beat rate (revenue and EPS separately) and the average / median next-day move, with the largest gap-up and gap-down.
  • The "beat and drop" count — quarters in which both lines beat and the stock fell. Three or more in eight is a pattern: this stock is graded on the guide or on a KPI, not on the headline.
  • Realized vs. implied — how often the actual move exceeded the options-implied move. This is the input for Phase 4.

Fewer than six quarters (recent IPO, spin-off) — say the sample is too thin for a base rate and lean on Phases 3 and 5 instead.

Phase 3 — What is priced in

Borrow the reverse-DCF logic from stock-valuation in a lighter form:

  1. Take the current price and a defensible forward multiple (the stock's own 3-year median, or the peer median from competitor-analysis).
  2. Back out the forward revenue / EPS the multiple requires over the next 12 months.
  3. Compare with consensus: is the price above what consensus supports (the market already assumes a beat-and-raise), at it, or below it (a plain beat is genuine news)?
  4. State the implied growth rate the price needs, in one line: "At $X and 32× forward, the price needs ~Y% EPS growth; consensus is Z%."

Add the positioning read where data exists: short interest and days-to-cover (short-interest), recent insider and institutional flow (insider-trading, institutional-ownership), and the stock's move over the last 30 days relative to its sector. A stock up 20% into the print with rising estimates has already had its beat; a stock down 15% with a stable consensus has room for relief.

Phase 4 — The implied move, and whether it is fairly priced

MeasureValueHow
Options-implied move (event only)±x.x%ATM straddle price ÷ stock price for the first expiry after the print. If that expiry is more than one session past the print, strip the non-event days so the horizon matches the next-day realized move: event move² ≈ near-straddle move² − (normal daily move)² × extra sessions, where the normal daily move comes from a longer-dated expiry's IV (IV × √(1/252)). State the horizon used
Realized average move (8 quarters)±x.x%from Phase 2
Realized median move±x.x%robust to one outlier quarter
Ratio: implied ÷ realized medianx.x> 1.2 the market is paying up for the event; < 0.8 it is complacent

Interpretation for a stockholder (not an options trader): the implied move is the market's estimate of the size of the reaction, in either direction — it says nothing about direction and is not a downside estimate. Model the miss-side downside separately in the scenario grid (the realized moves on past misses, the distance to the valuation floor from stock-valuation, the bear-case target). Then compare: if the larger of the implied move and the modelled miss-side downside exceeds the distance to the holder's stop or the bottom rung of the position-ladder, the position size is too large for the event — that is the finding, and it belongs in the grid's "miss" rule. Options strategy selection goes to options-analysis.

Phase 5 — The KPIs that will decide the reaction

List 3–5 company-specific numbers the market will read past the headline for — segment growth (data-center, cloud, subscribers), gross margin, unit economics (ARPU, take rate, deliveries), backlog / RPO, free cash flow, and the guidance range itself. For each: the consensus or prior value, the threshold that would count as good or bad, and where it will appear (press release vs. call). Then name the one number that has driven the stock in the last four prints. If a thesis-tracker file exists, its KPIs and thresholds take precedence — the print is that file's scheduled check.

Phase 6 — The scenario grid

Fill every cell before the print; leave nothing to be decided in the after-hours session.

ScenarioProbabilityTriggerExpected reactionPosition rule (decided now)
Beat & raise—Revenue and EPS ≥ whisper, guide midpoint above consensus+ implied move or more; fades if the stock ran into the printHold; add only if a position-ladder rung fills and the ceiling is not reached; never chase the gap
Beat & lower—Headline beat, guide midpoint below consensusFlat to − implied move; the most common "beat and drop"Trim into any early strength if the guide cut touches the thesis's mechanism; otherwise hold and re-run thesis-tracker --update
Miss—Revenue or EPS below consensus, or guide cut with no offsetting KPI− implied move or more; larger if positioning was longRespect the pre-set stop or the thesis-tracker exit trigger; do not average down inside the first two sessions
In line (state if material)—All three within ±1% of consensus, guide heldDrift, then trades on the call's toneNo action; run earnings-call-analysis on the transcript

Probabilities are the analyst's stated judgment, must sum to 100%, and must be justified by Phases 1–3 (a rising whisper lowers the "beat & raise" probability; a 15% drawdown into the print raises it). Where the user holds no position, the rules become entry rules: which scenario, at what price, would justify opening one — and which would not, regardless of price.

Phase 7 — What not to do

State these every time; they are the errors this skill exists to prevent:

  • Do not add size into the print on conviction. The print is a coin with a known payout distribution (Phase 4) and an unknown outcome; conviction does not change the distribution. Size for the miss case.
  • Do not hold undefined-risk short options through the event (short straddles, strangles, naked puts). The implied move is an estimate; the realized move has exceeded it in a measurable share of quarters (Phase 2). options-analysis covers defined-risk alternatives.
  • Do not react to the headline EPS before the guide and the KPI in Phase 5 are out. Most post-print reversals happen between the press release and the end of the call.
  • Do not treat a beat against a lowered bar as a beat. Compare with the 90-day-ago consensus (Phase 1), not only with today's.
  • Do not confuse the stock's reaction with the quality of the quarter. A drop on a good quarter is positioning; a rise on a poor one is relief. Log which one it was — thesis-tracker needs that distinction.

Phase 8 — Earnings Setup Score (0–10)

The headline number. It measures asymmetry for a holder — how much good news is already in the price versus how the stock has treated good news — not whether the print will be good.

ComponentWeight10 looks like0 looks like
Expectation bar (Phases 1, 3)35%Consensus flat or falling into the print; price implies less growth than consensus; whisper gap ≈ 0Consensus marked up sharply; price needs a beat-and-raise; whisper well above consensus
Beat history vs. reaction (Phase 2)25%High beat rate and positive median move on beats (the stock rewards good news)"Beat and drop" in ≥ 3 of 8 quarters
Event pricing (Phase 4)20%Implied move ≥ 1.2× realized median (downside is over-insured)Implied move < 0.8× realized median (complacency)
Positioning (Phase 3)10%Stock down and de-rated into the print; short interest elevated; no crowdingUp > 15% in 30 days with estimates rising; crowded long
Plan quality (Phases 5–6)10%A specific KPI threshold and a rule in every cellGeneric KPIs; rules left to "see how it trades"

A high score is not "buy before earnings." It means the setup is asymmetric in a holder's favour: the bar is low, the stock rewards beats, the downside is insured by the options market. A low score means the opposite — good news is largely priced and the stock has a habit of falling on beats — which for a holder argues for smaller size into the print, not for a short. Read the direction of the business from stock-eval; read the odds of the event from here.


3. Output Format

  1. Data & Sources header (above), including the confirmed date/time and the consensus as-of date
  2. Setup in three lines — the bar (consensus vs. whisper vs. prior guide), what the price assumes, and how this stock has treated beats
  3. Expectation stack table (Phase 1)
  4. 8-quarter history table and its three summary lines (Phase 2)
  5. What is priced in — the implied-growth sentence and the positioning read (Phase 3)
  6. Implied vs. realized move table and the position-size finding (Phase 4)
  7. KPIs to watch, with the one number that decides the reaction (Phase 5)
  8. Scenario grid with probabilities and position rules (Phase 6)
  9. What not to do (Phase 7)
  10. Thesis Invalidation and the Investment Signal block (below), with the Earnings Setup Score
  11. Hand-off line — "After the print: paste the release and transcript into earnings-call-analysis; then run thesis-tracker --update."

Example

User: earnings-preview NVDA — reports Aug 27 after the close. I hold 40 shares at $118;
      ladder ceiling is 60. Consensus and the last 8 quarters are pasted below. [pasted]

The assistant opens with the Data & Sources header (IR page for the date, pasted
consensus dated Aug 20, live straddle), then the setup in three lines: consensus has
been revised up 6% in 90 days and the whisper sits ~3% above it, so a consensus-level
beat will read as a miss; at the current price and a 3-year median forward multiple the
stock needs ~35% forward EPS growth against a consensus of ~30%, so the price already
assumes a beat-and-raise; the stock beat both lines in 8 of 8 quarters but fell the next
day in 3 of them — every time the guide midpoint was below the whisper.

The implied move is ±7.8% against a realized median of ±6.1% (ratio 1.28 — the event is
over-insured). The KPI that decides the reaction is data-center revenue growth versus
the ~40% whisper, then the next-quarter guide midpoint; gross margin is the tiebreaker.

Scenario grid: beat & raise 45% (hold; no add — the ladder's next rung is $105, far
below), beat & lower 35% (trim 10 shares into strength only if the guide cut names
demand, not supply), miss 20% (the exit trigger is the thesis-tracker file's ≥40% DC
growth KPI, not a price). What not to do: no adding into the print at 40 shares with a
60 ceiling and an implied move that reaches $112 on the downside.

Earnings Setup Score 4.6 / 10 — NEUTRAL: the bar is high and largely priced, but the
stock has rewarded beats and the downside is over-insured. Signal block, then the
hand-off to earnings-call-analysis and thesis-tracker --update.

Notes

  • Confirm the date from the company, not a calendar aggregator. Estimated dates are wrong often enough to matter — an implied move measured on the wrong expiry is meaningless.
  • Consensus is a moving target. Restate the as-of date next to every consensus figure; a number from three weeks ago is not the bar the stock will trade against.
  • Whisper is labeled an estimate every time it appears. A range with a rationale beats a false-precision point.
  • The score is about the event, not the company. A great business can have a poor setup (crowded, fully priced) and a mediocre one a good setup (washed out, low bar). Say which is which; do not blend them.
  • Position rules must respect position-ladder's ceiling and thesis-tracker's triggers when those exist. This skill never raises a ceiling or overrides an exit trigger.
  • After the print, close the loop. earnings-call-analysis on the transcript, then thesis-tracker --update, then record in its decision log which scenario occurred and whether the pre-set rule was followed. That log is what makes the next preview better.

Thesis Invalidation

After delivering the analysis signal, specify what would reverse it:

If signal is BULLISH (asymmetric setup for a holder) — the read breaks if:

  • Consensus or the whisper is revised up by more than ~3% (EPS) in the final week — the bar has moved and Phases 1 and 3 must be redone
  • The stock rallies more than the implied move before the print (a pre-announcement, a peer's blowout, a sector re-rating) — the relief has been taken in advance
  • A peer reports first and reveals the KPI in Phase 5 is deteriorating industry-wide

If signal is BEARISH (good news fully priced, poor beat history) — the read breaks if:

  • The stock sells off into the print by more than the implied move on no company news — the bar has been reset lower
  • Consensus is cut ahead of the print (the company or its peers guided down) — a beat against a lowered bar is not the same setup, and Phase 7's warning applies in reverse
  • Short interest rises sharply into the event — positioning has flipped and the reaction distribution with it

Re-run this analysis when:

  • The print itself — this analysis expires at the release; run earnings-call-analysis on the transcript and thesis-tracker --update on the numbers
  • Consensus, the guide, or the confirmed date changes
  • The stock moves more than half the implied move before the event
  • A direct peer reports
  • The options-implied move changes by more than 20% relative (repricing of the event)

Standard Signal Output

This skill measures the asymmetry of the earnings setup for a holder, not the direction of the business or the likely outcome of the print, so state the mapping explicitly: Signal: BULLISH means the bar is low, the stock rewards beats, and the downside is over-insured — the setup favours holding through the event at the current size; NEUTRAL means the setup is balanced or mixed — hold at current size, no adding; BEARISH means good news is largely priced and the stock has a habit of falling on beats — for a holder this argues for a smaller position into the print, not for a short. Action: BUY here means "the setup supports holding, and adding only via a pre-planned position-ladder rung"; Action: SELL means "reduce size into the event", never "short the print." Read the direction of the stock from stock-eval or bear-case, not from this block.

All analysis concludes with this standardized block:

In the output, the box below is followed immediately by this machine-readable footer — the same values as the box, filled in (no placeholders, no comments), so tools can read the result without parsing the box:

{"investskill": "1", "skill": "earnings-preview", "ticker": "<TICKER, or null for a portfolio / market-wide view>", "as_of": "<YYYY-MM-DD from Data & Sources>", "signal": "<BULLISH | NEUTRAL | BEARISH>", "confidence": "<HIGH | MEDIUM | LOW>", "horizon": "<SHORT-TERM | MEDIUM-TERM | LONG-TERM>", "score": 0.0, "action": "<BUY | HOLD | SELL>", "conviction": "<STRONG | MODERATE | WEAK>", "retrieval": "<pasted by user | web/tool retrieval | model memory>", "sources": ["<each primary source, dated>"]}
╔══════════════════════════════════════════════╗
║              INVESTMENT SIGNAL               ║
╠══════════════════════════════════════════════╣
║ Signal:      BULLISH / NEUTRAL / BEARISH     ║
║ Confidence:  HIGH / MEDIUM / LOW             ║
║ Horizon:     SHORT / MEDIUM / LONG-TERM      ║
║ Score:       X.X / 10                        ║
╠══════════════════════════════════════════════╣
║ Action:      BUY / HOLD / SELL               ║
║ Conviction:  STRONG / MODERATE / WEAK        ║
╚══════════════════════════════════════════════╝

Score Guide: 8.0–10.0 Strongly Bullish | 6.0–7.9 Moderately Bullish | 4.0–5.9 Neutral | 2.0–3.9 Moderately Bearish | 0.0–1.9 Strongly Bearish Confidence: HIGH (strong data, clear signals) | MEDIUM (mixed signals) | LOW (limited data, conflicting signals) Horizon: SHORT-TERM (1 week–3 months) | MEDIUM-TERM (3 months–1 year) | LONG-TERM (1+ years)

Note: The Score above is the Earnings Setup Score, mapped onto the standard scale for cross-skill comparability. It rates how asymmetric the event is for a holder — not how attractive the stock is and not the likelihood of a beat. Horizon is SHORT-TERM by construction: the analysis expires at the print. Confidence is capped at MEDIUM when the earnings date is unconfirmed, the whisper is not estimable, or fewer than six quarters of history exist.

Disclaimer: Educational analysis only. Not financial advice. Earnings outcomes are unknowable in advance; probabilities and scenario rules are a pre-commitment framework under stated assumptions, not a forecast or a trade instruction.

US Economics Analysis

⚠️ Data Verification — Do This Before Any Analysis

Before running any analysis, retrieve the latest release of every indicator you use:

  1. Fetch current readings — use web search or ask the user for the latest value and release date of each indicator (FRED, BLS, BEA, ISM, the Fed). Never assume a reading from training data; indicators are revised and re-released on a schedule.
  2. Confirm key figures — the policy rate, the 2s10s spread, the latest CPI/PCE, payrolls and unemployment, and PMI, each with its release date.
  3. State your data source — fill in the Data & Sources header (next section) so the origin, as-of date, retrieval path, and confidence of every figure are explicit at the top of the output.
  4. Flag stale data explicitly — if live data is unavailable, display this warning before proceeding:

⚠️ Live data unavailable. The following analysis uses training-data estimates which may be significantly out of date. Verify all prices and metrics before making any decisions.

Never silently substitute training-data estimates for current prices. When in doubt, ask the user to paste the latest release.


📋 Data & Sources Header — Open Every Output With It

The first thing in the output is this provenance block, filled in — never left as placeholders. It is the standard documented on the Data & Accuracy page and the first thing result-validator looks for:

Data & Sources
  As of:      <date the figures represent, e.g. 2026-06-30>
  Source:     <primary docs — SEC EDGAR 10-K/10-Q, company IR, FRED, exchange data …>
  Retrieval:  <pasted by user | web/tool retrieval | model memory>
  Confidence: <HIGH | MEDIUM | LOW>
  • Retrieval: model memory must be paired with Confidence: LOW — memory is a placeholder until confirmed against a primary source.
  • Mixed sources: list each with its own as-of date rather than blending them.
  • Sources disagree on the same figure (a closing price, a deal term): prefer the primary source (SEC filing, company release, exchange data) over an aggregator, then the most recent dated one. Show both values with their sources and dates, say which you used and why, and never average them.
  • Data the user pasted is reported as pasted by user; do not upgrade its confidence beyond what the user's own source supports.

🌐 Output Language — --lang zh-TW

English is the default. With --lang zh-TW — or when the user writes in Traditional Chinese and asks for a Chinese answer — write the whole analysis in Traditional Chinese (繁體中文, Taiwan usage), keeping it machine-checkable:

  • Translate every heading, narrative paragraph, table header and bullet. Give each metric's English name in brackets the first time it appears — 自由現金流(Free Cash Flow, FCF) — so the reader can match it to the glossary and to English analyses.
  • Keep in English tickers, figures and units, SEC form and item names (10-K Item 7, Form 4), the four Data & Sources field labels (As of, Source, Retrieval, Confidence — the Retrieval value may be written in Chinese, e.g. 使用者提供), and the Investment Signal box and JSON footer exactly as specified — field names and the BULLISH / NEUTRAL / BEARISH, HIGH / MEDIUM / LOW, BUY / HOLD / SELL, STRONG / MODERATE / WEAK values. Tools parse them.
  • Title the invalidation section 論點失效條件(Thesis Invalidation), and close with the disclaimer in Chinese: 免責聲明: 僅供教育用途,非投資建議。
  • Use Taiwan financial terms: 殖利率、本益比、營業利益率、自由現金流、聯準會、選擇權、買權/賣權、除息日、預扣稅.

Analyze US economic conditions and their implications for investment decisions.

Key Economic Indicators

  1. Growth Indicators

    • GDP growth rate and components
    • Employment data (NFP, unemployment rate, jobless claims)
    • Consumer spending and retail sales
    • Manufacturing and services PMI
  2. Inflation Metrics

    • CPI (Consumer Price Index)
    • PCE (Personal Consumption Expenditures)
    • PPI (Producer Price Index)
    • Wage growth trends
  3. Monetary Policy

    • Federal Reserve policy stance
    • Interest rates (Fed Funds rate, Treasury yields)
    • Money supply and bank lending
    • Fed meeting minutes and forward guidance
  4. Market Sentiment

    • Consumer confidence indices
    • Business sentiment surveys
    • Credit spreads and risk indicators
    • Market volatility (VIX)
  5. Fiscal Policy

    • Government spending and stimulus programs
    • Tax policy changes
    • Budget deficit and debt levels

Analysis Framework

  • Identify current economic cycle phase
  • Assess policy implications for different sectors
  • Evaluate recession/expansion risks
  • Determine impact on equity, bond, and commodity markets
  • Provide sector rotation recommendations

Yield Curve Analysis

Key Spreads to Monitor

SpreadDefinitionCurrent1-Year Avg10-Year AvgSignal
2s10s10yr Treasury minus 2yr Treasury
3M10Y10yr Treasury minus 3-Month T-Bill
5s30s30yr Treasury minus 5yr Treasury

3M10Y is the historically strongest recession predictor (NY Fed model is based on this spread).

Yield Curve Shapes

ShapeDescriptionEconomic Implication
Normal (Steep)Long-term rates well above short-term ratesHealthy growth expectations, bank margins expanding
FlatShort and long-term rates near parityLate-cycle signal, growth slowing, Fed near peak
InvertedShort-term rates above long-term ratesRecession warning — markets pricing in rate cuts ahead
Bear SteepeningBoth ends rise, long end rises fasterInflation concern, term premium expanding
Bull SteepeningBoth ends fall, short end falls fasterCutting cycle underway, growth relief expected

Inversion Duration and Recession Lead Time

Historical precedent for 3M10Y inversion:

Inversion DurationHistorical Recession Lead Time
< 3 monthsUnreliable signal
3–6 months12–18 months typically
6–12 months6–15 months typically
> 12 monthsHigh confidence; within 12 months

Rule of thumb: Yield curve uninversion (re-steepening after inversion) is often the more immediate warning — recession tends to arrive shortly after the curve re-steepens from inversion.

Fed Rate Cycle Positioning

  • Hiking Cycle: Fed raising rates — short end rises faster, curve flattens/inverts. Growth stocks under pressure.
  • Pause: Fed on hold — curve stabilizes. Markets watch for pivot signals.
  • Cutting Cycle: Fed reducing rates — short end falls faster, curve steepens. Risk-on environment, cyclicals and growth stocks benefit.

Real Yields (TIPS) Analysis

  • Real Yield = Nominal Treasury Yield − Breakeven Inflation Rate (derived from TIPS)
  • Rising real yields: Tighten financial conditions. Negative for long-duration assets (growth stocks, gold, long bonds).
  • Falling real yields: Easier financial conditions. Positive for growth stocks, gold, emerging markets, long bonds.
  • 10-Year Real Yield thresholds: Below 0% is historically accommodative; above 2% is meaningfully restrictive.
  • Breakeven inflation (5-year, 5-year forward): Market's long-run inflation expectation. Persistently above 2.5% signals inflation concern.

Credit Market Indicators

Investment Grade (IG) Credit Spreads (OAS — Option-Adjusted Spread)

Spread LevelConditionInterpretation
< 100 bpsNormalRisk appetite healthy, credit markets functioning
100–150 bpsCautionStress emerging, watch for tightening conditions
> 150 bpsStressCredit markets seizing, risk-off, watch equities

High Yield (HY) Credit Spreads

Spread LevelConditionInterpretation
< 350 bpsNormalBenign default environment, strong risk appetite
350–500 bpsCautionElevated risk aversion, avoid lower-quality credits
> 500 bpsDistressRecession/financial stress scenario, significant HY risk
> 800 bpsCrisisSystemic credit event risk, similar to 2008/2020 episodes

Rule: HY spreads lead equity markets by 2–4 weeks on average. Widening HY spreads while equities hold = warning signal.

TED Spread

  • Definition: 3-Month LIBOR (now SOFR) minus 3-Month T-Bill yield
  • Measures interbank lending stress and counterparty risk appetite in the banking system
  • Normal: < 50 bps
  • Elevated stress: 50–100 bps
  • Crisis signal: > 100 bps (peaked at ~450 bps during 2008 GFC)

MOVE Index (Bond Market Volatility)

  • Bond market equivalent of VIX — measures implied volatility in US Treasury options
  • Normal: 80–100
  • Elevated: 100–130 (policy uncertainty, high rate volatility)
  • Crisis: > 150 (1994, 2008, 2020, 2023 banking crisis)
  • High MOVE compresses equity valuations by increasing discount rates unpredictably.

Credit as a Leading Indicator

  • IG/HY spread widening before equity weakness is a leading warning (credit sees risk first)
  • Spread compression while equities lag = catch-up potential, constructive signal
  • IG vs. HY divergence: If HY widens but IG holds, idiosyncratic credit stress — watch lower-quality equities
  • Leveraged loan market: CLO issuance and leveraged loan spreads reflect private credit conditions

Global Macro Comparison

Economic Cycle Positioning (US vs. EU vs. China)

EconomyCurrent PhaseGDP GrowthInflationPolicy StanceEquity Implication
United States
Eurozone
China
Japan
UK

Economic cycle phases: Early Expansion → Mid Expansion → Late Expansion → Contraction → Recovery

PMI Comparison Across Major Economies

Country/RegionPMI IndexLast ReadingTrendAbove/Below 50
USISM Mfg
USISM Services
EurozoneMarkit Mfg
EurozoneMarkit Svcs
ChinaCaixin Mfg
ChinaOfficial PMI

Rule: PMI above 50 = expansion; below 50 = contraction. Composite PMI below 48 for 2+ months is recessionary signal.

Central Bank Divergence Analysis

Central BankCurrent RateLast MoveNext Expected MoveCycle Phase
Federal Reserve (Fed)
European Central Bank (ECB)
Bank of Japan (BOJ)
Bank of England (BOE)
People's Bank of China (PBOC)

Divergence signals:

  • Fed tightening while ECB/BOJ easing → USD strengthens, EM currencies weaken
  • Synchronized easing → Global risk-on, EM outperforms, commodities bid
  • BOJ policy normalization → JPY strengthens, unwinds carry trades

Dollar (DXY) Strength and Sector Impact

DXY DirectionUS Multinational EarningsCommoditiesEmerging MarketsDomestic US Small-Caps
Strengthening (rising DXY)Headwind (FX translation)BearishBearish (USD-denominated debt stress)Relative outperform
Weakening (falling DXY)TailwindBullishBullishRelative underperform
  • DXY above 105: Meaningful headwind for S&P 500 multinationals (roughly 40% of S&P revenues are foreign)
  • DXY below 95: Significant tailwind, boosts international earnings in USD terms

Emerging Market Vulnerability Indicators

  • EM FX pressure: Current account deficits + elevated external USD debt = vulnerable to dollar strength
  • EM Debt Stress Index: Sovereign spread widening in EM bonds (EMBI+ spread)
  • Capital outflow risks: Rate differential between US and EM narrows during Fed cutting cycles — can reverse
  • China contagion risk: Property sector stress, credit impulse, and stimulus effectiveness
  • Commodity-exporting EMs: Benefit from commodity supercycles; inversely, hurt by USD strength

Recession Probability Scoring

New York Fed Recession Model

Based on the 3M10Y yield curve spread, the NY Fed publishes a monthly recession probability for the next 12 months.

Probability RangeInterpretation
0–10%Expansion — very low recession risk
10–25%Low risk — monitor indicators
25–50%Elevated — caution warranted
50–75%High risk — recession likely within 12 months
> 75%Near-certain — defensive positioning required

Current NY Fed reading: ____%

Conference Board Leading Economic Index (LEI)

The LEI composite combines 10 leading indicators across financial markets, labor, manufacturing, and consumer expectations.

  • Consecutive monthly declines (3+): Strong recession warning
  • Year-over-year decline > 4%: Historically aligned with recessions
  • LEI component breakdown: Manufacturing hours, building permits, consumer expectations, credit spread, yield curve, stock prices, initial jobless claims

Current LEI trend: Rising / Flat / Declining

Sahm Rule

Sahm Rule Indicator = Current 3-month average unemployment rate minus the minimum of the 3-month average unemployment rate over the prior 12 months.

  • Threshold: ≥ 0.5 percentage points = Real-time recession signal with high historical accuracy
  • Triggered in every US recession since 1970 — and once without one so far (August 2024), so a trigger raises the odds rather than confirming a recession
  • Works in real-time without revision lag that affects other indicators

Current Sahm Indicator reading: ____

Custom Composite Recession Probability

Scoring model combining: Yield curve signal + LEI trend + Sahm Rule + Credit spreads + PMI momentum

ZoneScore RangeInterpretation
Expansion0–25%Risk-on appropriate; cyclicals, growth outperform
Caution25–50%Balanced positioning; reduce cyclical overweights
High Risk50–75%Defensive rotation; increase quality, reduce leverage
Near-Certain75–100%Full defensive posture; cash, defensives, short vol

Historical recession episodes and leading indicators:

RecessionYield Curve InversionLEI DeclineSahm TriggerS&P 500 Peak-to-Trough
2001 (Dot-com)2000YesYes−49%
2008 (GFC)2006–2007YesYes−57%
2020 (COVID)2019YesYes−34%
2022–2024 (no recession declared)2022–2024 (longest on record)YesYes (Aug 2024)−25% (2022 bear market)

The 2022–2024 row is the counter-example: the curve inverted for about two years, the LEI fell, and the Sahm rule triggered in August 2024, yet no recession had been declared as of this writing. Treat the indicators as shifts in probability, not verdicts, and say so in the output.


Output

Deliver concise economic assessment with:

  • Current economic state summary
  • Key risks and opportunities
  • Sector and asset class implications
  • Investment positioning recommendations

Thesis Invalidation

After delivering the analysis signal, specify what would reverse it:

If signal is BULLISH — thesis breaks if:

  • yield curve inverts >50bps AND leading indicators fall for 3 consecutive months
  • [One or two more triggers drawn from this analysis's own drivers, each with a threshold]

If signal is BEARISH — thesis breaks if:

  • yield curve normalizes AND PMI recovers above 52 for 2+ months
  • [One or two more triggers drawn from this analysis's own drivers, each with a threshold]

Re-run this analysis when:

  • Next earnings release
  • Price moves ±15% from current level
  • 60 days have elapsed
  • Material news event (acquisition, leadership change, regulatory decision)

Standard Signal Output

All analysis concludes with this standardized block:

In the output, the box below is followed immediately by this machine-readable footer — the same values as the box, filled in (no placeholders, no comments), so tools can read the result without parsing the box:

{"investskill": "1", "skill": "economics-analysis", "ticker": "<TICKER, or null for a portfolio / market-wide view>", "as_of": "<YYYY-MM-DD from Data & Sources>", "signal": "<BULLISH | NEUTRAL | BEARISH>", "confidence": "<HIGH | MEDIUM | LOW>", "horizon": "<SHORT-TERM | MEDIUM-TERM | LONG-TERM>", "score": 0.0, "action": "<BUY | HOLD | SELL>", "conviction": "<STRONG | MODERATE | WEAK>", "retrieval": "<pasted by user | web/tool retrieval | model memory>", "sources": ["<each primary source, dated>"]}
╔══════════════════════════════════════════════╗
║              INVESTMENT SIGNAL               ║
╠══════════════════════════════════════════════╣
║ Signal:      BULLISH / NEUTRAL / BEARISH     ║
║ Confidence:  HIGH / MEDIUM / LOW             ║
║ Horizon:     SHORT / MEDIUM / LONG-TERM      ║
║ Score:       X.X / 10                        ║
╠══════════════════════════════════════════════╣
║ Action:      BUY / HOLD / SELL               ║
║ Conviction:  STRONG / MODERATE / WEAK        ║
╚══════════════════════════════════════════════╝

Score Guide: 8.0–10.0 Strongly Bullish | 6.0–7.9 Moderately Bullish | 4.0–5.9 Neutral | 2.0–3.9 Moderately Bearish | 0.0–1.9 Strongly Bearish Confidence: HIGH (strong data, clear signals) | MEDIUM (mixed signals) | LOW (limited data, conflicting signals) Horizon: SHORT-TERM (1 week–3 months) | MEDIUM-TERM (3 months–1 year) | LONG-TERM (1+ years)

Disclaimer: Educational analysis only. Not financial advice.

ETF Analysis — What You Actually Own, and What It Costs You

⚠️ Data Verification — Do This Before Any Analysis

Before running any analysis, always retrieve the latest fund data for the ticker:

  1. Fetch current price and NAV — use web search or ask the user for the live price, the fund's net asset value, and the premium/discount to NAV. Never assume a price from training data.
  2. Confirm the fund facts from the issuer — expense ratio, index tracked, AUM, inception date, holdings count, top-10 weights, and distribution history come from the issuer's factsheet and prospectus, not from memory. Fund facts change: expense ratios get cut, indexes get swapped, funds get merged or closed.
  3. State your data source — fill in the Data & Sources header (next section) so the origin, as-of date, retrieval path, and confidence of every figure are explicit at the top of the output.
  4. Flag stale data explicitly — if live data is unavailable, display this warning before proceeding:

⚠️ Live data unavailable. The following analysis uses training-data estimates which may be significantly out of date. Verify all prices and metrics before making any decisions.

Never silently substitute training-data estimates for current fund facts. When in doubt, ask the user to paste the factsheet.


📋 Data & Sources Header — Open Every Output With It

The first thing in the output is this provenance block, filled in — never left as placeholders. It is the standard documented on the Data & Accuracy page and the first thing result-validator looks for:

Data & Sources
  As of:      <date the figures represent, e.g. 2026-06-30>
  Source:     <primary docs — SEC EDGAR 10-K/10-Q, company IR, FRED, exchange data …>
  Retrieval:  <pasted by user | web/tool retrieval | model memory>
  Confidence: <HIGH | MEDIUM | LOW>
  • Retrieval: model memory must be paired with Confidence: LOW — memory is a placeholder until confirmed against a primary source.
  • Mixed sources: list each with its own as-of date rather than blending them.
  • Sources disagree on the same figure (a closing price, a deal term): prefer the primary source (SEC filing, company release, exchange data) over an aggregator, then the most recent dated one. Show both values with their sources and dates, say which you used and why, and never average them.
  • Data the user pasted is reported as pasted by user; do not upgrade its confidence beyond what the user's own source supports.

🌐 Output Language — --lang zh-TW

English is the default. With --lang zh-TW — or when the user writes in Traditional Chinese and asks for a Chinese answer — write the whole analysis in Traditional Chinese (繁體中文, Taiwan usage), keeping it machine-checkable:

  • Translate every heading, narrative paragraph, table header and bullet. Give each metric's English name in brackets the first time it appears — 自由現金流(Free Cash Flow, FCF) — so the reader can match it to the glossary and to English analyses.
  • Keep in English tickers, figures and units, SEC form and item names (10-K Item 7, Form 4), the four Data & Sources field labels (As of, Source, Retrieval, Confidence — the Retrieval value may be written in Chinese, e.g. 使用者提供), and the Investment Signal box and JSON footer exactly as specified — field names and the BULLISH / NEUTRAL / BEARISH, HIGH / MEDIUM / LOW, BUY / HOLD / SELL, STRONG / MODERATE / WEAK values. Tools parse them.
  • Title the invalidation section 論點失效條件(Thesis Invalidation), and close with the disclaimer in Chinese: 免責聲明: 僅供教育用途,非投資建議。
  • Use Taiwan financial terms: 殖利率、本益比、營業利益率、自由現金流、聯準會、選擇權、買權/賣權、除息日、預扣稅.

Overview

Most portfolios are built on an ETF core, yet every other framework in this catalog evaluates a single company. Nobody asks the questions that decide whether an ETF is a good vehicle: what does it cost to hold, does it deliver the index it promises, can you get out of it, what is actually inside it, and how much of it do you already own through other positions?

This skill is due diligence on the wrapper, not a market call. It does not predict whether US large caps will outperform; sector-analysis and economics-analysis do that. It answers: given that you want this exposure, is this fund a fit way to get it — and is it a fit for the portfolio you already have? The output is an ETF Fitness Score (0–10) built from cost, tracking, liquidity, what-you-own, overlap, distributions, and structure, plus a direct comparison against two or three peers and against simply buying the top five holdings.

Where it sits. stock-screener ranks stocks; portfolio-review judges the whole portfolio; sector-analysis picks where to be. This skill sits between them: it vets the fund that would carry the exposure, then hands the result to portfolio-review (does it fit the allocation?), risk-stress-test (what does it do in a drawdown?), and tax-lens (what does the wrapper cost after tax?).


1. Inputs

InputRequiredDefault if unstated
ETF ticker(s) — one fund, or several to compare✅—
The exposure the user thinks they are buying ("US large cap", "global ex-US", "dividend growth")recommendedinfer from the fund name and index, and say so
Current holdings with weights (tickers, other funds)optional — required for the overlap sectionoverlap reported as "not assessed" and its 15% weight redistributed (Phase 9)
Intended purchase size (as % of the post-purchase portfolio)optional — required for post-purchase exposure10% of the portfolio, stated
Account type (taxable / IRA / non-US)optionaltaxable US assumed; state it
Holding horizonoptional5+ years

Overlap and tilt need holdings data. Top-10 weights are on every factsheet; the full holdings file (issuer CSV) is needed for an exact overlap or factor tilt. If the user has not pasted it, compute from the top 10 only, label the result as a lower bound, and say what file would sharpen it.


2. Framework

Phase 1 — Fund card

Establish the facts before any judgment:

FieldWhy it matters
Issuer, ticker, inception dateFunds under 3 years old have no full-cycle record; funds under $100M AUM have closure risk
Index tracked (or "active")The index is the strategy — read its methodology: cap-weighted? screened? rebalanced when?
Expense ratio (net, after waivers) and waiver expiryA fee waiver that expires is a fee increase with a date
Structure: physical replication / sampling / synthetic (swap) / ETN / grantor trust / commodity poolDetermines counterparty risk, tax form, and whether "holdings" means anything
Legal domicile (US 1940-Act ETF, Irish UCITS, …)Drives withholding tax and estate-tax exposure for non-US holders
Distribution policy (accumulating / distributing, frequency)Matters for taxable accounts and for income users

Phase 2 — Cost and tracking

Expense ratio vs. category. Compare to the median for funds tracking the same kind of index, not to "all ETFs":

CategoryCheapExpensiveNeeds a reason
Broad US large-cap index≤ 0.05%> 0.20%> 0.75%
Broad international developed / EM index≤ 0.10%> 0.35%> 0.75%
Sector, factor, dividend, thematic≤ 0.15%> 0.50%> 0.95%
Active, leveraged, inverse, alternatives≤ 0.50%> 1.00%— (the whole product needs a reason)

Tracking difference vs. tracking error — keep them apart.

  • Tracking difference = fund total return − index total return over a period. This is the cost you actually paid. For a well-run physical fund it should be close to −(expense ratio); a fund that trails its index by more than the expense ratio every year is leaking money (sampling, cash drag, poor securities lending, withholding tax at the fund level).
  • Tracking error = the standard deviation of that difference. It measures consistency, not cost. Low tracking error with a large negative tracking difference is a fund that reliably underperforms.

Report the 1-, 3-, and 5-year tracking difference from the issuer's own performance table (NAV return vs. index return). Flag a tracking gap whenever |tracking difference| exceeds the expense ratio by more than 0.10% per year — and say which direction (a positive gap usually means securities-lending revenue, which is fine but not guaranteed).

Phase 3 — Liquidity and closure risk

MetricComfortableCautionWarning
AUM> $1B$100M – $1B< $100M (closure risk; issuers cull small funds)
Average daily $ volume> $10M$1M – $10M< $1M (use limit orders, avoid the open/close)
Bid-ask spread (median, %)≤ 0.05%0.05% – 0.25%> 0.25% (a round trip costs more than a year of fees)
Premium/discount to NAV (typical range)±0.10%±0.50%wider, or persistently one-sided

Note the caveat that matters: ETF liquidity comes from the underlying basket, not only from the ETF's own volume — a thinly traded ETF on liquid large caps trades fine with limit orders; a heavily traded ETF on illiquid bonds does not. Say which case applies.

Phase 4 — What you own

  • Holdings count and top-10 weight. Flag concentration when top-10 > 40% for a "broad" fund, or when a single name > 10%. Cap-weighted indexes concentrate in bull markets — a fund that was diversified five years ago may not be now.
  • Sector tilt — the fund's sector weights vs. the broad benchmark the user thinks they are buying (e.g. the total US market). Report the three largest over/under-weights in percentage points.
  • Factor tilt — size, value/growth, quality, momentum, low-vol, yield: which way does the index methodology lean, and does the user know it? A "dividend" fund is usually a value + low-growth tilt; an "equal-weight" fund is a small-cap tilt.
  • Country / currency exposure for international funds — and whether the fund is currency-hedged (a bet on the dollar, held whether the user wants it or not).
  • Look-through: if the fund holds other funds, look through to the final securities.

Phase 5 — Overlap with the user's other holdings

With the holdings the user supplied, compute overlap by weight:

  1. Look through every existing fund position to its underlying securities (top holdings from the issuer's factsheet; label the coverage — "top 10 = 34% of the fund" — so the reader knows the rest is unassessed). Existing single stocks are their own exposure.
  2. Express both sides as weights of the same denominator: the candidate ETF's holdings as % of the ETF; the existing portfolio's look-through exposures as % of the existing portfolio.
  3. Overlap % = Σ over shared securities of min(weight in ETF, weight in existing portfolio). This is the share of the ETF that duplicates what the user already owns, on a like-for-like basis.
  4. With the intended purchase size p (as % of the post-purchase portfolio; default 10%, stated), compute the post-purchase exposure to each shared name: (1 − p) × existing weight + p × ETF weight.

Report:

  • Overlap % (by weight) and the duplicated names, largest first; the look-through coverage used.
  • The effective post-purchase exposure to the top duplicated names (e.g. "at a 10% purchase you would own the largest holding through three vehicles at a combined 11% of the portfolio").
  • A plain verdict: redundant (> 60% overlap — this is the same bet in a second wrapper), partially overlapping (25–60% — fine if intentional), complementary (< 25%).

Without a holdings file, report the top-10-only overlap as a lower bound and say so.

Phase 6 — Distributions and taxes

  • Trailing 12-month distribution yield and its composition: qualified dividends, non-qualified income, return of capital, capital gains. REIT-, bond-, and commodity-heavy funds pay mostly non-qualified income taxed at ordinary rates.
  • Capital-gains distributions, last 5 years. Most equity ETFs distribute none, because in-kind creation/redemption lets them hand appreciated shares to authorized participants instead of selling them; a mutual fund cannot. An equity ETF that has distributed gains (heavy rebalancing, index reconstitution, small AUM with big redemptions) deserves a note. Bond and commodity funds are a different matter.
  • Tax form: 1099 for ordinary ETFs; K-1 and 60/40 treatment for many commodity and volatility funds structured as partnerships; grantor-trust rules for physically backed metals. A K-1 is a paperwork and timing cost, not just a tax rate.
  • Hand the detail to tax-lens — this skill flags, that one computes.

Phase 7 — Structure warnings

State any that apply, in a box the reader cannot miss:

StructureWarning to print
Leveraged (2×, 3×) or inverseResets daily. Over more than a day, path dependence ("volatility decay") means the return is not 2× or −1× the index. A holding tool for days, not years — say so even if the user asked for a long-term view
Synthetic / swap-basedThe fund holds a swap, not the securities. Counterparty exposure to the swap provider; "holdings" listed are collateral
ETNAn unsecured note issued by a bank. If the issuer fails, the ETN fails, regardless of the index. Check the issuer's credit and whether the ETN is still creating shares (closed ETNs trade at wild premiums)
Single-stock ETFA leveraged or option-overlaid position in one company. Concentration + decay + fees — nearly always dominated by owning the stock
Commodity pool / futures-basedRoll yield (contango) can make the fund diverge sharply from spot; K-1 tax form; not an investment in the physical commodity
Currency-hedgedEmbeds a short position in the foreign currency; hedging cost ≈ the interest-rate differential
Covered-call / buffer / defined-outcomeCaps upside by design; the "yield" is option premium, partly return of your own capital. Compare total return, never headline yield
Actively managedCompare to the cheapest index fund in the same category over the manager's tenure, not to the category average

Phase 8 — Alternatives and the direct-ownership check

  • Peers: two or three funds tracking the same or a very similar index. Compare expense ratio, 3-yr tracking difference, AUM, spread, and structure in one table. Prefer the cheapest fund with the smallest tracking gap and adequate liquidity — everything else is marketing.
  • ETF vs. buying the top 5 holdings directly: what share of the fund's weight the top 5 represent, the fund's fee on that share, what the investor would give up (the other N−5 holdings, rebalancing, dividend reinvestment convenience) and what they would gain (zero ongoing fee, tax-lot control, no tracking gap). For a broad fund with top-5 weight under 25%, direct ownership is a different, more concentrated bet — say so. For a "sector" fund where the top 5 are 60%+ of the weight, direct ownership is a serious alternative.

Phase 9 — ETF Fitness Score (0–10)

The headline number. It measures fitness of the vehicle for the stated exposure and portfolio, not the attractiveness of the asset class.

ComponentWeight10 looks like0 looks like
Cost & tracking25%Cheapest quartile for its category; tracking difference ≈ −ER with no gapExpensive for the category and a tracking gap > 0.25%/yr
Liquidity & viability15%AUM > $1B, spread ≤ 0.05%, tight to NAVAUM < $100M or spread > 0.25% or persistent discount
What you own vs. what you wanted20%Index methodology matches the stated exposure; tilts disclosed and acceptableThe fund is a different bet than the user thinks (hidden factor/country/currency tilt, top-10 > 50%)
Overlap with existing holdings15%< 25% overlap, or overlap is intentional> 60% overlap — a duplicate position
Distribution & tax efficiency10%No capital-gains distributions; income mostly qualified; 1099Regular gains distributions; K-1; non-qualified income in a taxable account
Structure15%Plain physical replicationLeveraged/inverse, ETN, or single-stock held as a long-term position

Hard caps: a leveraged/inverse/single-stock fund evaluated for a horizon longer than a few weeks caps at 3.0; an ETN caps at 5.0; > 60% overlap caps at 5.0. State any cap that fired.

When overlap is not assessed (no holdings supplied): drop the component and renormalize the remaining five weights to 100% (Cost & tracking 29.4% · Liquidity 17.6% · What you own 23.5% · Distribution 11.8% · Structure 17.6%), print Overlap: not assessed — supply holdings for the full score under the score, and cap Confidence at MEDIUM. Never score an unassessed component as 10 or as 0.


3. Output Format

Data & Sources · as of · issuer factsheet · fund prospectus · retrieval · confidence

1. Fund card          ticker · issuer · index tracked · inception · AUM · expense ratio · structure · domicile
2. Cost & tracking    expense ratio vs. category median · 1/3/5-yr tracking difference · tracking gap flag · spread · ADV
3. What you own       #holdings · top-10 weight · sector / factor / country / currency tilt · concentration flag
4. Overlap            % overlap by weight with the user's holdings · duplicated names · verdict (redundant / partial / complementary)
5. Distributions      trailing yield · composition (qualified / ordinary / ROC / gains) · capital-gains distributions (last 5 yrs) · tax form
6. Structure risks    leveraged / inverse / synthetic / ETN / single-stock / commodity / hedged / covered-call → warning box
7. Alternatives       2–3 peers in one table · "vs. top-5 holdings directly" table
8. ETF Fitness Score  0–10 with the six sub-scores and any cap that fired → Thesis Invalidation → signal block

When several tickers are given, produce sections 1–6 per fund compactly, then one comparison table, then one score per fund.


Example

User: Is VOO or SPYM the better core holding? I already hold VTI at 40% of my portfolio.

The assistant builds the fund card for both (same index family — S&P 500 — so the
comparison collapses to cost, tracking, liquidity and share price), pulls the issuer
performance tables to compare 1/3/5-yr tracking difference against each fund's expense
ratio, notes that both are physical, distributing, 1099 funds with no capital-gains
distributions in the record supplied, and that either is liquid enough for any retail
order with a limit. Then it runs the overlap check against VTI: an S&P 500 fund is
roughly four-fifths of a total-US-market fund by weight, so the overlap is high and the
verdict is "redundant — you already own this exposure through VTI; adding either fund
raises large-cap concentration without adding a new bet." Both funds score well on
fitness in isolation; the report's headline is the overlap finding, and the suggested
next step is `portfolio-review` to decide whether the user actually wants *more* US
large cap, or `etf-analysis` on an international or small-cap fund instead.

Notes

  • This is vehicle due diligence, not a market view. Do not score the asset class. A perfectly built fund on an exposure the user should not own still gets a high Fitness Score — pair with sector-analysis or economics-analysis for whether to own it, and portfolio-review for how much.
  • Never invent fund statistics. Expense ratios, AUM, tracking figures, and holdings change; if the issuer page or factsheet was not retrieved or pasted, say "not verified" rather than recalling a number.
  • Read the index methodology. Two funds with similar names can track very different indexes (screened vs. unscreened, capped vs. uncapped, quarterly vs. annual rebalancing). The methodology document, not the fund name, is the strategy.
  • Non-US investors (including readers using the Traditional Chinese site): a US-domiciled ETF exposes a non-resident holder to 30% (or treaty-rate) withholding on distributions and to US estate-tax exposure above a low exemption; an Irish-domiciled UCITS equivalent typically suffers 15% withholding at the fund level and no US estate-tax exposure, at a somewhat higher expense ratio and, for accumulating share classes, no distributions at all. Name the UCITS equivalent when one exists and hand the arithmetic to tax-lens --non-us.
  • Overlap arithmetic depends on the data supplied. Top-10-only overlap is a lower bound; label it. Full holdings files are downloadable from every major issuer.
  • Bond, commodity, and alternative ETFs need the extra questions this skill only flags (duration, credit quality, roll yield, collateral). Say when the fund is outside the equity-index case this framework is built for.
  • Pairs with: portfolio-review (allocation fit), risk-stress-test (drawdown behaviour), tax-lens (after-tax cost), sector-analysis / stock-screener (whether the exposure is worth owning at all).

Thesis Invalidation

After delivering the analysis signal, specify what would reverse it:

If signal is BULLISH (the fund is a fit vehicle) — the verdict breaks if:

  • The tracking difference widens beyond −(expense ratio + 0.10%) in the next annual performance table — the fund has started leaking
  • The issuer changes the index, raises the fee, lets a waiver expire, or announces a merger or liquidation (AUM falling below ~$50M is the early warning)
  • The fund's concentration crosses the thresholds above (top-10 > 40%, single name > 10%) so that "broad" is no longer true
  • The user's other holdings change such that the overlap verdict moves from complementary to redundant

If signal is BEARISH (the fund is not a fit vehicle) — the verdict breaks if:

  • A fee cut or a fixed tracking gap removes the cost objection (re-run against the peers)
  • The user's stated exposure changes so that the tilt this fund carries is now the intended bet
  • The overlap objection disappears because the duplicated position is sold
  • For a structure warning: the user's horizon and purpose genuinely match the product (a multi-day hedge with an inverse fund, for example) — the warning still prints, the cap may be lifted

Re-run this analysis when:

  • The issuer publishes its annual performance / tracking table
  • Any change to expense ratio, index, or fund structure is announced
  • The user adds or removes a position large enough to change the overlap verdict
  • AUM falls below $100M or the spread widens persistently
  • 12 months have elapsed

Standard Signal Output

This skill measures vehicle fitness, not market direction, so state the mapping: Signal: BULLISH means the fund is a fit way to hold the stated exposure for this portfolio; NEUTRAL means it is acceptable but a cheaper, better-tracking, or less overlapping alternative exists (name it); BEARISH means do not use this wrapper (structure cap fired, redundant overlap, or a tracking gap that peers do not have). Action: BUY here means "this fund is fit for the purpose" — it is not a view on the asset class. Read whether to own the exposure from sector-analysis or portfolio-review, not from this block.

All analysis concludes with this standardized block:

In the output, the box below is followed immediately by this machine-readable footer — the same values as the box, filled in (no placeholders, no comments), so tools can read the result without parsing the box:

{"investskill": "1", "skill": "etf-analysis", "ticker": "<TICKER, or null for a portfolio / market-wide view>", "as_of": "<YYYY-MM-DD from Data & Sources>", "signal": "<BULLISH | NEUTRAL | BEARISH>", "confidence": "<HIGH | MEDIUM | LOW>", "horizon": "<SHORT-TERM | MEDIUM-TERM | LONG-TERM>", "score": 0.0, "action": "<BUY | HOLD | SELL>", "conviction": "<STRONG | MODERATE | WEAK>", "retrieval": "<pasted by user | web/tool retrieval | model memory>", "sources": ["<each primary source, dated>"]}
╔══════════════════════════════════════════════╗
║              INVESTMENT SIGNAL               ║
╠══════════════════════════════════════════════╣
║ Signal:      BULLISH / NEUTRAL / BEARISH     ║
║ Confidence:  HIGH / MEDIUM / LOW             ║
║ Horizon:     SHORT / MEDIUM / LONG-TERM      ║
║ Score:       X.X / 10                        ║
╠══════════════════════════════════════════════╣
║ Action:      BUY / HOLD / SELL               ║
║ Conviction:  STRONG / MODERATE / WEAK        ║
╚══════════════════════════════════════════════╝

Score Guide: 8.0–10.0 Strongly Bullish | 6.0–7.9 Moderately Bullish | 4.0–5.9 Neutral | 2.0–3.9 Moderately Bearish | 0.0–1.9 Strongly Bearish Confidence: HIGH (strong data, clear signals) | MEDIUM (mixed signals) | LOW (limited data, conflicting signals) Horizon: SHORT-TERM (1 week–3 months) | MEDIUM-TERM (3 months–1 year) | LONG-TERM (1+ years)

Note: The Score above is the ETF Fitness Score, mapped onto the standard scale for cross-skill comparability. It rates how fit this fund is as a vehicle for the stated exposure and portfolio — not how attractive the exposure is. Confidence is LOW whenever the issuer factsheet was not retrieved or the overlap was computed from the top 10 only.

Disclaimer: Educational analysis only. Not financial advice. Fund facts change; verify every figure against the issuer's current factsheet and prospectus before acting.

Fact Check — Verify the Numbers, Cite the Sources

⚠️ Data Verification — Do This Before Any Analysis

Before verifying anything, establish what data each claim actually needs:

  1. Fetch current price only where a claim needs it — market claims (price, market cap, yield, multiples "today") are checked against a live quote for the report's as-of date and today's; use web search or ask the user. Reported figures, historical comparisons, and --recompute runs need no live quote — do not request one, and do not let its absence block them. Never assume a price from training data.
  2. Confirm key figures — for this skill the "key figures" are the sources themselves: the filing (10-K / 10-Q / 8-K / proxy / Form 4 / 13F), the IR release, the FRED series, the exchange or issuer data, or the document the user pasted. Retrieve or request them before judging a single claim. A claim checked against memory is not checked.
  3. State your data source — fill in the Data & Sources header (next section) so the origin, as-of date, retrieval path, and confidence of every figure are explicit at the top of the output.
  4. Flag stale data explicitly — if a live quote is needed for a market claim and unavailable, display this warning before proceeding, and mark those claims ❓ or 🕒 rather than guessing:

⚠️ Live data unavailable. The following analysis uses training-data estimates which may be significantly out of date. Verify all prices and metrics before making any decisions.

The warning describes the inputs actually used: when every claim was checked against user-supplied documents, say so instead of implying training-data estimates were involved. Never silently substitute training-data estimates for current prices. When in doubt, ask the user to paste the latest quote.


📋 Data & Sources Header — Open Every Output With It

The first thing in the output is this provenance block, filled in — never left as placeholders. It is the standard documented on the Data & Accuracy page and the first thing result-validator looks for:

Data & Sources
  As of:      <date the figures represent, e.g. 2026-06-30>
  Source:     <primary docs — SEC EDGAR 10-K/10-Q, company IR, FRED, exchange data …>
  Retrieval:  <pasted by user | web/tool retrieval | model memory>
  Confidence: <HIGH | MEDIUM | LOW>
  • Retrieval: model memory must be paired with Confidence: LOW — memory is a placeholder until confirmed against a primary source.
  • Mixed sources: list each with its own as-of date rather than blending them.
  • Sources disagree on the same figure (a closing price, a deal term): prefer the primary source (SEC filing, company release, exchange data) over an aggregator, then the most recent dated one. Show both values with their sources and dates, say which you used and why, and never average them.
  • Data the user pasted is reported as pasted by user; do not upgrade its confidence beyond what the user's own source supports.

For this skill the header describes the sources used to verify, not the report being verified — the report's own header is quoted and judged in Phase 5.


🌐 Output Language — --lang zh-TW

English is the default. With --lang zh-TW — or when the user writes in Traditional Chinese and asks for a Chinese answer — write the whole analysis in Traditional Chinese (繁體中文, Taiwan usage), keeping it machine-checkable:

  • Translate every heading, narrative paragraph, table header and bullet. Give each metric's English name in brackets the first time it appears — 自由現金流(Free Cash Flow, FCF) — so the reader can match it to the glossary and to English analyses.
  • Keep in English tickers, figures and units, SEC form and item names (10-K Item 7, Form 4), the four Data & Sources field labels (As of, Source, Retrieval, Confidence — the Retrieval value may be written in Chinese, e.g. 使用者提供), and the Investment Signal box and JSON footer exactly as specified — field names and the BULLISH / NEUTRAL / BEARISH, HIGH / MEDIUM / LOW, BUY / HOLD / SELL, STRONG / MODERATE / WEAK values. Tools parse them.
  • Title the invalidation section 論點失效條件(Thesis Invalidation), and close with the disclaimer in Chinese: 免責聲明: 僅供教育用途,非投資建議。
  • Use Taiwan financial terms: 殖利率、本益比、營業利益率、自由現金流、聯準會、選擇權、買權/賣權、除息日、預扣稅.

Overview

Every InvestSkill analysis opens with a Data & Sources header, and result-validator scores whether sources are "cited or identifiable". Neither answers the question a careful reader actually has: is this specific number true, and where exactly does it come from? A report can carry an honest header, a tidy methodology, and a consistent signal — and still be built on a revenue figure from the wrong fiscal year, a margin computed on non-GAAP earnings, or a share count that was stale when the model wrote it.

This skill closes that gap at the level of the individual claim. It takes any report — an InvestSkill output, a broker note, a spreadsheet, a paragraph from a forum post — pulls out every figure and factual assertion, checks each one against a primary source, recomputes every derived number from verified inputs, and hands back three things: a claim ledger with a verdict per claim (✅ verified · ⚠️ mismatch · ❓ unverifiable · 🔁 recomputed · 🕒 stale), a corrected copy of the report with inline citations and a References section, and a Verification Score 0–10 that result-validator can consume as the evidence behind its Data Quality dimension.

Division of labour. result-validator asks is this analysis well built? (methodology, consistency, risk coverage, transparency). fact-check asks are the inputs true? Run this one first when the stakes are real; the validator's score means little if the numbers underneath it are wrong. learning-coach then explains the result to a beginner.

What it is not. It does not re-run the analysis, form its own view on the stock, or change the report's signal. It never invents a source to make a claim look verified: an unverifiable claim stays ❓, and a claim that cannot be traced is more honestly left uncited than given a plausible-looking reference. Fabricating a citation is the one failure mode that makes this skill worse than useless — it is a hard-cap-to-zero offence in the score.


1. Modes

ModeInvocationWhat it does
Verify (default)paste the report (+ sources, or permission to retrieve them)Full ledger, recomputation, consistency check, corrected report with citations, Verification Score
Cite--cite-only + the report + its sourcesAttach inline citations and a References section to figures the sources confirm; no recomputation, no score beyond coverage — for a report you already trust but need to make traceable
Recompute--recompute + the reportOnly the arithmetic: rebuild every derived figure from the reported inputs and flag disagreements — fast, needs no external source
Diff--diff + two versions of a report or two data setsWhich figures changed, by how much, and whether the change is a restatement, a period roll, or an error

If a source cannot be retrieved in the environment, the skill says which claims that blocks and asks the user to paste the document — it never fills the gap from memory.

2. Inputs

InputRequiredDefault if unstated
The report or data to check (pasted, or a file)✅—
Sources: pasted filing text / IR release / data table, URLs, or permission to retrieve✅ for Verify and Citeask; without sources every claim is ❓ and the score is capped
Ticker and fiscal calendar (FY end month)recommendedinfer from the report; state it — half of all "mismatches" are fiscal-year confusions
Tolerance for market data (price, market cap)optional±1% and same trading day
Which claims are signal-driving (the 3–5 numbers the report's conclusion rests on)optionalinfer from the report's own signal box and thesis; state the choice
Output form: corrected copy vs. annotated diffoptionalcorrected copy with strike-through of originals

3. Framework

Phase 1 — Extract every claim into a ledger

Read the report once and list every checkable statement — not just the headline numbers. One row per claim:

#Claim (as written)Value · unit · periodTypeSignal-driving?Where in report
1"Revenue grew 20% to $6.12B in FY2026"$6,120M · FY2026 · +20% YoYreported (filing)✅§2 Financial Health
2"FCF of $930M"$930M · FY2026derived (OCF − capex)✅§2
3"trades at 24× forward earnings"24.2× · as of ?market × estimate✅§3 Valuation
4"CFO bought $1.2M of stock in May"$1.2M · 2026-05fact (Form 4)—§5 Insider
5"top customer is 18% of revenue"18% · FY2026reported (10-K risk factors)—§1

Claim types decide which source can verify them: reported (a figure stated in a filing or release), derived (computed from reported figures — must be recomputed, not just matched), market (price, market cap, yield — needs an as-of date), estimate (consensus, guidance — needs the source and date of the estimate), macro (rates, CPI — FRED / BLS series and vintage), fact (an event: a filing, a transaction, an appointment), qualitative (a characterization — "dominant", "improving" — checked for whether the cited evidence supports it, never marked verified on its own).

Mark the signal-driving claims. They carry more weight in the score and a mismatch on one of them changes the report's standing.

Phase 2 — Source hierarchy

Only a source the reader could open verifies a claim. Rank what you use, and record the tier in the ledger:

TierSourcesVerifies
1 — PrimarySEC EDGAR filings (10-K, 10-Q, 8-K, DEF 14A, Form 4, 13F, S-1), the company's own IR release or presentation, FRED / BLS / BEA series, the exchange's or ETF issuer's published data, IRS publications, the fund prospectus / factsheetreported figures, facts, macro, fund data
1 — User-suppliedThe document the user pasted (a filing excerpt, a broker statement, a data export)its own figures — labelled user-supplied so the reader knows verification stops at the user's document
2 — SecondaryReputable financial data vendors and press explicitly quoting a primary document, consensus aggregators for estimatesmarket data and estimates, with the vendor and date recorded; a reported figure verified only at Tier 2 is marked ✅ (secondary)
3 — Not a sourceModel memory, forum posts, uncited summaries, a second AI outputnothing — a claim matched only against these stays ❓

Prefer the most specific location: Item and Note number, statement line, table, page. "The 10-K" is a document, not a citation.

Retrieving a Tier-1 source without an API key. When the user grants permission to retrieve and the host can fetch URLs, go straight to SEC EDGAR rather than a summary site: resolve the CIK from https://www.sec.gov/files/company_tickers.json (zero-pad cik_str to 10 digits); list filings from https://data.sec.gov/submissions/CIK##########.json, where filings.recent.form / filingDate / reportDate / accessionNumber / primaryDocument line up by index (10-K, 10-Q, 8-K, DEF 14A, 4 for insider transactions — Form 4s are filed under the issuer's CIK as well as the insider's — and 20-F / 6-K for foreign issuers); open the document at https://www.sec.gov/Archives/edgar/data/<CIK without leading zeros>/<accession without dashes>/<primaryDocument>; and cross-check statement figures against the tagged facts at https://data.sec.gov/api/xbrl/companyfacts/CIK##########.json. Send an identifying User-Agent and stay under 10 requests per second. Record Retrieval: web/tool retrieval, the accession number and the filing date. Institutional holdings are the exception to the issuer-CIK route: Form 13F-HR is filed by the investment manager, so it never appears under the target company's CIK. Resolve the manager (from the claim: "Vanguard holds 8.9%") to its CIK, list its 13F-HR filings, and verify the position in the information table by the target's CUSIP — or use EDGAR full-text search (https://www.sec.gov/edgar/search/#/q=%22<company or CUSIP>%22&forms=13F-HR) to find which managers report the security. A 13F verifies one manager's position at one quarter-end; an aggregate "institutional ownership %" is a secondary-source figure and is marked ✅ (secondary) at best. If the host cannot fetch, ask the user to paste the document — or to run node scripts/fetch-edgar.js <TICKER> --form 10-K / node scripts/fetch-fundamentals.js <TICKER> from the InvestSkill repository and paste the output (Retrieval: pasted by user). A source that could not be retrieved leaves its claims ❓; it is never filled from memory.

Phase 3 — Verify each claim

For each ledger row, locate the figure in the source, compare, and assign one verdict:

VerdictMeaningRule
✅ VERIFIEDThe source states the figureReported figures: exact, or rounding within 0.5%. Market data: same trading day and within the stated tolerance. Percentages: within 0.1 pp. Record source · location · as-of
⚠️ MISMATCHThe source states a different figureRecord the source value, the delta (absolute and %), and the likely cause — the usual suspects: wrong fiscal year or quarter; TTM vs. fiscal year; GAAP vs. adjusted; basic vs. diluted shares; a restated prior period; a currency or unit slip (thousands vs. millions); a stale price
❓ UNVERIFIABLENo Tier 1–2 source availableSay what source would settle it. This is not "false" — never let the corrected report read as if the claim were disproven
🔁 RECOMPUTEDA derived figure rebuilt from verified inputsShow the arithmetic (Phase 4); mark agrees or disagrees (report X, recomputed Y)
🕒 STALEVerified, but the as-of date is older than the report implies or than 90 daysVerified figure, wrong vintage — give the current value alongside

A qualitative claim is judged only on whether the evidence the report cites for it holds; it is never ✅ on its own.

Phase 4 — Recompute every derived figure

Never accept a derived number by matching it to a secondary source that may have made the same mistake. Rebuild it from verified inputs and show the work:

Free cash flow      = OCF − capex                          1,240 − 310            = 930   ✓ report 930
Gross margin        = gross profit ÷ revenue               3,550 ÷ 6,120          = 58.0% ✓ report 58%
Revenue growth      = rev / prior rev − 1                  6,120 ÷ 5,100 − 1      = 20.0% ✓
Net debt            = total debt − cash                    800 − 1,650            = −850  (net cash) ✓
Market cap          = price × shares outstanding           48.20 × 412            = 19,858   (shares outstanding at the price date — not the diluted count used for EPS)
Enterprise value    = market cap + debt − cash             19,858 + 800 − 1,650   = 19,008
P/E (trailing)      = price ÷ diluted EPS                  48.20 ÷ 1.99           = 24.2× ✓ report "24× forward" ✗ — trailing, not forward
FCF yield           = FCF ÷ market cap                     930 ÷ 19,858           = 4.7%
Payout (FCF basis)  = dividends paid ÷ FCF                 165 ÷ 930              = 17.7%

Standard checks: FCF, margins, growth rates, per-share figures (which share count?), P/E · P/S · EV/EBITDA · P/FCF (trailing vs. forward, and whose estimate), net debt and EV, payout and yield, ROE / ROIC (state the denominator), any score or weighted composite in the report (do the weights sum to 100%? does the weighted sum match?). Where the report's definition is ambiguous, say so rather than picking one silently.

Phase 5 — Consistency and honesty of the report itself

  • Numbers vs. narrative — does the text say what the tables say? ("margins expanding" next to a table where they fell)
  • Internal contradictions — the same figure quoted two ways in two sections.
  • Signal vs. score — the signal box's Score must sit in the band its Signal implies (≥ 6.0 BULLISH · 4.0–5.9 NEUTRAL · < 4.0 BEARISH); a mismatch is a red flag, not a rounding issue.
  • The report's Data & Sources header vs. what was verifiable — a header claiming Retrieval: web/tool retrieval · Confidence: HIGH over a report whose figures cannot be traced is itself a finding. Quote the header and state whether the verification supports it.
  • Fabricated or unopenable citations in the report — a reference that does not exist, or a page/Item that does not contain the figure, is the most serious finding this skill can make. Say so plainly.

Phase 6 — Re-issue the report with citations

Produce the corrected copy (or the diff, if asked):

  • Every ✅ figure gets an inline citation [n]; every 🔁 agrees figure gets the citation of its inputs plus (recomputed).
  • Every ⚠️ figure is corrected in place, with the original struck through: $5.9B $6.12B [1] — and a one-line note on the cause. Never silently change a number.
  • Every ❓ figure is marked [?] and left as written. Every 🕒 figure keeps its citation and gains (as of <date>; current: <value> [m]).
  • Qualitative claims whose evidence failed are marked [unsupported].
  • A References section closes the report:
## References
[1] Zephyr Robotics Inc., Form 10-K for FY ended 2026-03-31 — Item 8, Consolidated Statements of Operations, p. 54 (SEC EDGAR, accession 0000000000-26-000000; retrieved 2026-06-30)
[2] Same filing — Item 8, Consolidated Statements of Cash Flows, p. 58
[3] Zephyr Robotics Q4 FY2026 earnings release, 2026-05-14 — segment table (company IR)
[4] Form 4, filed 2026-05-19 — CFO open-market purchase (SEC EDGAR)
[5] User-supplied: price and shares outstanding as of 2026-06-30 close (pasted by user; not independently verified)

Each reference names the document, the exact location, the as-of or filing date, and how it was retrieved. If the environment cannot produce a URL or accession number, say so rather than inventing one.

Phase 7 — Verification Score (0–10)

The headline number measures how much of the report is demonstrably true, not whether the report's conclusion is right.

ComponentWeight10 looks like0 looks like
Coverage30%≥ 95% of claims ✅ or 🔁 agrees< 40%
Signal-driving accuracy30%Every signal-driving claim verified within toleranceA signal-driving claim off by > 5% or from the wrong period
Recomputation15%Every derived figure agreesDerived figures disagree or cannot be rebuilt from the report's own inputs
Source quality15%All Tier 1, specific locationsMostly secondary or user-supplied only
Freshness10%Every as-of date matches the report's claim and is < 90 days old for market dataStale figures presented as current

Hard caps (state any that fires): a signal-driving ⚠️ or 🔁 disagrees with delta > 5% → max 4.0 · more than half the claims ❓ → max 5.0 · the report's own Data & Sources header overstates its retrieval or confidence → max 6.0 · any fabricated or unopenable citation found in the report → 0.0. Always print a Caps fired: line under the score (none, or the list) so a capped 4.0 is never mistaken for an earned 4.0 — a report whose score was capped by a signal-driving mismatch is a failed verification regardless of the number.

Map onto the standard bands: ≥ 6.0 the report's inputs are sound (BULLISH on the report) · 4.0–5.9 usable with the listed corrections (NEUTRAL) · < 4.0 do not rely on it — re-run the analysis with verified inputs (BEARISH). Hand the score and the ledger to result-validator, whose Data Quality dimension should not exceed what this score supports.


4. Output Format

  1. Data & Sources header — the sources used to verify, with retrieval path
  2. Verification summary — one line: N claims · V verified · R recomputed (r agree) · M mismatch · U unverifiable · S stale · Verification Score X.X · Caps fired: none | <list>
  3. Signal-driving claims first — the 3–5 that matter, each with verdict, source, and delta
  4. Claim ledger — the full table from Phase 1 with verdict, source tier, location, as-of
  5. Mismatches in detail — original · source value · delta · likely cause · what the corrected figure does to the report's conclusion
  6. Recomputation table — Phase 4, arithmetic shown
  7. Report consistency findings — Phase 5, including the judgment on the report's own header
  8. Corrected report with citations (or --diff output) and the References section
  9. Thesis Invalidation — what would change these verdicts (below)
  10. Investment Signal block — mirrors the report's signal with the verification verdict attached (below)

Example

User: [pastes a stock-eval output on ZEPH that concludes BULLISH 7.4 / 10] Verify every number
      and add sources. Here is the FY2026 10-K excerpt and the Q4 release. [pasted]

The assistant lists 23 claims, flags 5 as signal-driving (revenue growth, gross margin, FCF,
net cash, forward P/E). Verifies 17 against the 10-K and the release with Item/statement/page
locations; recomputes 6 derived figures — all agree except one: the report's "24× forward P/E"
is 48.20 ÷ 1.99, i.e. the *trailing* multiple on FY2026 diluted EPS; no forward estimate is
cited, so the claim becomes ⚠️ MISMATCH (label) with the corrected wording "24.2× trailing";
consensus forward EPS would be needed for the forward figure → ❓ until supplied. Two claims
are 🕒 STALE: the price and market cap are from a date 40 days before the report. One
qualitative claim ("dominant share of warehouse robotics") cites no evidence → [unsupported].
The report's own header said "Retrieval: web/tool retrieval · Confidence: HIGH"; 17/23 verified
supports MEDIUM, so that is noted. No fabricated citations. Verification Score 6.8 — the report's
inputs are sound; the signal stands with the valuation sentence corrected. Emits the corrected
report with [1]–[9] inline and a References section, and hands the ledger to result-validator.

Notes

  • Run it before result-validator, not instead of it. This skill checks the inputs; the validator checks the reasoning built on them. The validator's Data Quality score should never exceed what the Verification Score supports.
  • Every InvestSkill report is a candidate, but so is anything else the user brings — a broker note, a newsletter, a spreadsheet, a screenshot transcribed to text. The ledger discipline is the same.
  • Unverifiable is not false. The corrected copy must never read as if a ❓ claim were refuted. The honest output for a claim without a source is the claim, unchanged, with [?].
  • Fiscal calendars cause most false mismatches. Establish the fiscal year end and the period each table covers before comparing anything.
  • Do not fabricate a reference to fill a gap. No accession number, page, or URL that was not actually seen. If the environment cannot retrieve, ask the user to paste.
  • Pairs with: result-validator (consumes the score), learning-coach (explains the ledger to a beginner), 10k-digest and financial-report-analyst (produce well-cited inputs in the first place), full-report (run this on the composite before shipping it), thesis-tracker (verify the KPI values before recording them at a check).

Thesis Invalidation

After delivering the analysis signal, specify what would reverse it:

If signal is BULLISH (the report's inputs verified) — the verdicts break if:

  • A new filing restates a figure the report relied on (an 8-K or an amended 10-K/A) — every ✅ on that figure becomes 🕒 until re-checked
  • A signal-driving figure that was verified at Tier 2 (secondary) is later contradicted by the primary document
  • The market-data as-of dates pass 90 days — verified prices become stale prices, and the valuation claims built on them lose their ✅

If signal is BEARISH (the report's inputs did not verify) — the verdicts break if:

  • The user supplies the source that turns the ❓ claims into ✅ — the ledger is re-run, not amended by hand
  • A mismatch is traced to a period or definition difference the report actually disclosed elsewhere (then it is a labelling issue, not a wrong number — downgrade to a note)
  • The report is re-issued by its author with the corrected figures, in which case fact-check the new version, not the old

Re-run this analysis when:

  • The company files its next 10-Q / 10-K, or any 8-K restating prior figures
  • The report is edited or re-generated
  • 90 days pass on any market-data claim
  • Before the report is shared, published, or used to size a position
  • result-validator scores Data Quality above what the ledger supports

Standard Signal Output

This skill does not form a view on the stock. Score is always the Verification Score, and Confidence is capped at the tier it supports (≥ 8.0 HIGH · 6.0–7.9 MEDIUM · < 6.0 LOW). The other fields depend on the outcome:

  • Verification passed (Score ≥ 4.0 and no signal-driving cap fired — Caps fired: none, or only the coverage / header caps): Signal, Horizon, Action, and Conviction mirror the verified report's own, and a line under the box states Mirrors the report's signal — quoted, not re-derived. If a signal-driving claim mismatched by more than 5%, add Signal not supported at stated confidence — re-run the analysis with verified inputs.
  • Verification failed (Score < 4.0, or the signal-driving mismatch cap fired — even if the capped score reads exactly 4.0 — or a fabricated citation → 0): the box does not carry the report's Signal or Action forward — an unsupported BUY must not survive into the verification result. Fill Signal: NEUTRAL · Action: HOLD · Conviction: WEAK · Confidence: LOW, and under the box write Verification failed — the report's own signal (BULLISH / BUY, quoted) is not supported by its inputs; do not act on it.
  • For a report with no signal block, fill Signal / Action / Conviction with n/a.

All analysis concludes with this standardized block:

In the output, the box below is followed immediately by this machine-readable footer — the same values as the box, filled in (no placeholders, no comments), so tools can read the result without parsing the box:

{"investskill": "1", "skill": "fact-check", "ticker": "<TICKER, or null for a portfolio / market-wide view>", "as_of": "<YYYY-MM-DD from Data & Sources>", "signal": "<BULLISH | NEUTRAL | BEARISH>", "confidence": "<HIGH | MEDIUM | LOW>", "horizon": "<SHORT-TERM | MEDIUM-TERM | LONG-TERM>", "score": 0.0, "action": "<BUY | HOLD | SELL>", "conviction": "<STRONG | MODERATE | WEAK>", "retrieval": "<pasted by user | web/tool retrieval | model memory>", "sources": ["<each primary source, dated>"]}
╔══════════════════════════════════════════════╗
║              INVESTMENT SIGNAL               ║
╠══════════════════════════════════════════════╣
║ Signal:      BULLISH / NEUTRAL / BEARISH     ║
║ Confidence:  HIGH / MEDIUM / LOW             ║
║ Horizon:     SHORT / MEDIUM / LONG-TERM      ║
║ Score:       X.X / 10                        ║
╠══════════════════════════════════════════════╣
║ Action:      BUY / HOLD / SELL               ║
║ Conviction:  STRONG / MODERATE / WEAK        ║
╚══════════════════════════════════════════════╝

Score Guide: 8.0–10.0 Strongly Bullish | 6.0–7.9 Moderately Bullish | 4.0–5.9 Neutral | 2.0–3.9 Moderately Bearish | 0.0–1.9 Strongly Bearish Confidence: HIGH (strong data, clear signals) | MEDIUM (mixed signals) | LOW (limited data, conflicting signals) Horizon: SHORT-TERM (1 week–3 months) | MEDIUM-TERM (3 months–1 year) | LONG-TERM (1+ years)

Note: The Score above is the Verification Score — how much of the report is demonstrably true — mapped onto the standard scale for cross-skill comparability. When verification passes, Signal, Horizon, Action, and Conviction are the verified report's own, quoted unchanged; when it fails, they are replaced by NEUTRAL / HOLD / WEAK so that no unsupported call is carried forward.

Disclaimer: Educational analysis only. Not financial advice. Verification is limited to the sources available at run time; an unverifiable claim is neither confirmed nor refuted.

Financial Report Analyst

⚠️ Data Verification — Do This Before Any Analysis

Before running any analysis, always retrieve the latest market data for the ticker:

  1. Fetch current price — use web search or ask the user for the live price, 52-week range, and market cap. Never assume a price from training data.
  2. Confirm key figures — the filing itself (form, fiscal period, filing date, accession number) and its headline figures — revenue, operating income, net income, OCF, capex, cash, debt, diluted shares — read from the statements, not a data site.
  3. State your data source — fill in the Data & Sources header (next section) so the origin, as-of date, retrieval path, and confidence of every figure are explicit at the top of the output.
  4. Flag stale data explicitly — if live data is unavailable, display this warning before proceeding:

⚠️ Live data unavailable. The following analysis uses training-data estimates which may be significantly out of date. Verify all prices and metrics before making any decisions.

Never silently substitute training-data estimates for current prices. When in doubt, ask the user to paste the latest quote.


📋 Data & Sources Header — Open Every Output With It

The first thing in the output is this provenance block, filled in — never left as placeholders. It is the standard documented on the Data & Accuracy page and the first thing result-validator looks for:

Data & Sources
  As of:      <date the figures represent, e.g. 2026-06-30>
  Source:     <primary docs — SEC EDGAR 10-K/10-Q, company IR, FRED, exchange data …>
  Retrieval:  <pasted by user | web/tool retrieval | model memory>
  Confidence: <HIGH | MEDIUM | LOW>
  • Retrieval: model memory must be paired with Confidence: LOW — memory is a placeholder until confirmed against a primary source.
  • Mixed sources: list each with its own as-of date rather than blending them.
  • Sources disagree on the same figure (a closing price, a deal term): prefer the primary source (SEC filing, company release, exchange data) over an aggregator, then the most recent dated one. Show both values with their sources and dates, say which you used and why, and never average them.
  • Data the user pasted is reported as pasted by user; do not upgrade its confidence beyond what the user's own source supports.

🌐 Output Language — --lang zh-TW

English is the default. With --lang zh-TW — or when the user writes in Traditional Chinese and asks for a Chinese answer — write the whole analysis in Traditional Chinese (繁體中文, Taiwan usage), keeping it machine-checkable:

  • Translate every heading, narrative paragraph, table header and bullet. Give each metric's English name in brackets the first time it appears — 自由現金流(Free Cash Flow, FCF) — so the reader can match it to the glossary and to English analyses.
  • Keep in English tickers, figures and units, SEC form and item names (10-K Item 7, Form 4), the four Data & Sources field labels (As of, Source, Retrieval, Confidence — the Retrieval value may be written in Chinese, e.g. 使用者提供), and the Investment Signal box and JSON footer exactly as specified — field names and the BULLISH / NEUTRAL / BEARISH, HIGH / MEDIUM / LOW, BUY / HOLD / SELL, STRONG / MODERATE / WEAK values. Tools parse them.
  • Title the invalidation section 論點失效條件(Thesis Invalidation), and close with the disclaimer in Chinese: 免責聲明: 僅供教育用途,非投資建議。
  • Use Taiwan financial terms: 殖利率、本益比、營業利益率、自由現金流、聯準會、選擇權、買權/賣權、除息日、預扣稅.

Extract deep investment insights from official company financial filings and reports — 10-K annual reports, 10-Q quarterly reports, earnings press releases, proxy statements (DEF 14A), and investor presentations. Combines quantitative financial analysis with qualitative assessment of management tone, strategy, and risk disclosure.

Overview

Official filings are the primary source of truth for company analysis. Unlike financial data terminals (which aggregate selected metrics), the actual 10-K and 10-Q contain management's own words, full footnote disclosures, legal risk factors, and accounting policy choices that are invisible in summary data feeds. This skill teaches you to read between the lines of official documents and convert dense regulatory language into actionable investment insights.


How to Use

Provide the report in one of these ways:

# Paste report text directly
/financial-report-analyst [paste 10-K or 10-Q sections]

# Reference a specific ticker for the most recent filing
/financial-report-analyst AAPL 10-K

# Analyze a specific section
/financial-report-analyst MSFT 10-Q --section risk-factors

# Full deep-dive analysis
/financial-report-analyst NVDA 10-K --full

# Comparative: compare current vs prior period filing
/financial-report-analyst GOOGL --compare-prior

Retrieving the filing — the keyless EDGAR path

When the user names a ticker instead of pasting text, this skill still fetches nothing by itself. If the host can fetch URLs, get the primary document from SEC EDGAR (free, no API key) and mark Retrieval: web/tool retrieval:

  1. Ticker → CIK: https://www.sec.gov/files/company_tickers.json — match ticker (BRK-B for BRK.B), zero-pad cik_str to 10 digits.
  2. Filings index: https://data.sec.gov/submissions/CIK##########.json — in filings.recent, the arrays form · filingDate · reportDate · accessionNumber · primaryDocument line up by index. Pick the newest row whose form is the one asked for: 10-K (annual; 20-F for foreign private issuers), 10-Q (quarterly; 6-K for foreign issuers), 8-K (earnings release is usually Exhibit 99.1 in the filing folder), DEF 14A (proxy).
  3. Document: https://www.sec.gov/Archives/edgar/data/<CIK without leading zeros>/<accessionNumber without dashes>/<primaryDocument> — the full filing as HTML. The folder's <accessionNumber>-index.htm lists every exhibit.
  4. Statement facts (to cross-check Item 8): https://data.sec.gov/api/xbrl/companyfacts/CIK##########.json.

Identify yourself with a User-Agent header (Name email) and stay under 10 requests per second. For --compare-prior, take the second-newest row of the same form. Cite accession number and filing date in the Data & Sources header.

If the host cannot fetch, ask the user to paste the sections — or, from the InvestSkill repository, to run node scripts/fetch-edgar.js <TICKER> --form 10-Q (saves the filing as text plus a .json with the header fields) or node scripts/fetch-fundamentals.js <TICKER> (a reconciled statement pack from the XBRL facts) — and mark Retrieval: pasted by user. With neither, proceed only under the ⚠️ warning above with Confidence: LOW.


Sector KPI Packs (--sector)

Generic operating-company metrics mislead in five sectors. When the user passes --sector <bank|insurer|reit|biotech|saas> — or the company is clearly in one of them (SIC code, business description) — swap the generic metrics for the sector pack, say which pack was applied, and drop the metrics in its "Ignore" line (or show them only with a note that they are not meaningful here). Pull the KPIs from the 10-K / 10-Q and, where they are not GAAP line items (CET1, occupancy, NRR), from the company's supplemental disclosures, regulatory filings or earnings release — never from memory.

/stock-eval ZEPH --sector saas
/financial-report-analyst ZEPH 10-K --sector reit

bank

  • Swap in: CET1 ratio vs. its regulatory minimum + buffers · net interest margin (NIM) · efficiency ratio (non-interest expense / revenue) · NPL ratio and net charge-off (NCO) ratio · allowance / loans · loan-to-deposit ratio · deposit beta (Δ deposit cost / Δ policy rate) and uninsured-deposit share · AOCI and held-to-maturity unrealised losses vs. tangible equity · ROTCE · P/TBV
  • Ignore: P/E as the primary multiple, EV/EBITDA, Altman Z, Beneish, current ratio, DSO/DIO, debt/equity and FCF (deposits and loans are the business, not its financing)
  • Read: efficiency < 60% is good; NCO ratio against the bank's own cycle history; HTM losses above ~25% of tangible equity are a warning; justify P/TBV with ROTCE (fair P/TBV ≈ ROTCE / cost of equity)

insurer

  • Swap in: combined ratio = loss ratio + expense ratio (P&C) · prior-year reserve development (favourable / adverse, % of opening reserves) · net investment yield and portfolio credit mix · book value per share growth (with and ex-AOCI) · ROE · P/B · solvency: RBC ratio (US) or Solvency II ratio
  • Ignore: EV/EBITDA, Altman Z, Beneish, FCF, working-capital days, and premium growth as a quality signal (fast growth can mean under-pricing)
  • Read: combined ratio < 100% = underwriting profit; repeated adverse development is a major flag; BVPS growth plus dividends is the compounding measure; justify P/B with sustainable ROE

reit

  • Swap in: FFO and AFFO per share (NAREIT FFO; AFFO deducts recurring capex, straight-line rent and leasing costs) · P/FFO and P/AFFO · occupancy · same-store NOI growth · weighted average lease expiry (WALE) and the lease-expiry ladder · net debt / EBITDA and fixed-charge coverage · AFFO payout ratio · NAV and implied cap rate
  • Ignore: EPS and P/E (real-estate depreciation distorts GAAP earnings), Altman X2 (REITs must distribute income), FCF before AFFO adjustments
  • Read: AFFO payout < ~85% leaves room to grow the dividend, > 100% is unfunded; sector net debt / EBITDA typically runs 5–7×; falling occupancy or negative same-store NOI is the operating warning

biotech

  • Swap in: cash runway in quarters = (cash + investments) / quarterly operating burn · pipeline table (asset, indication, phase, next readout, partner, economics) · probability of success to approval by phase · catalysts (readouts, PDUFA dates, advisory committees) · partnerships (upfronts, milestones, royalties, who holds which rights) · dilution history (shares outstanding over 3–5 years, ATM use, offerings and their discounts)
  • Ignore: P/E, EV/EBITDA, margins, ROIC, Piotroski F-Score (pre-revenue companies fail it by construction), and DCFs built on management's peak-sales claims
  • Read: runway < 6 quarters means a raise is near. Rough base rates to approval from industry clinical-success studies (BIO / Informa, all indications): Phase 1 ≈ 8–10%, Phase 2 ≈ 15%, Phase 3 ≈ 50–60%, filed (NDA/BLA) ≈ 90%; oncology runs lower. Risk-adjust every pipeline value by them and hand the dates to catalyst-calendar

saas

  • Swap in: ARR and ARR growth · net revenue retention (NRR) and gross revenue retention (GRR) · Rule of 40 (revenue growth % + FCF margin %) · CAC payback (months) = S&M / (net-new ARR × gross margin) × 12 · subscription gross margin · SBC-adjusted FCF margin · RPO and current RPO (cRPO) growth vs. revenue growth
  • Ignore: GAAP P/E for early-stage names, EV/EBITDA that adds back SBC, headline FCF margin without deducting SBC, current ratio (deferred revenue inflates current liabilities)
  • Read: NRR > 110% strong, < 100% means the base is shrinking; GRR > 90%; Rule of 40 ≥ 40 on the SBC-adjusted margin; CAC payback < 24 months; cRPO growing slower than revenue is a leading deceleration signal

How the pack changes the rest of the analysis. Valuation switches to the sector multiple (P/TBV, P/B, P/AFFO, risk-adjusted pipeline value, EV/ARR or EV/revenue vs. growth); quality uses the sector KPIs in place of the generic margin and ratio checks; risk adds the sector's failure mode (deposit flight and rate marks for banks, reserve inadequacy for insurers, refinancing and occupancy for REITs, trial failure and dilution for biotech, retention decay for SaaS). Everything else — data verification, thesis invalidation and the signal block — is unchanged. State in the output which generic metrics were dropped and why.


Step-by-Step Analysis Framework

Phase 1: Document Orientation

Before diving into numbers, understand what kind of document this is and what period it covers:

  • Filing Type: 10-K (annual), 10-Q (quarterly), 8-K (material event), DEF 14A (proxy), S-1 (IPO)
  • Fiscal Period: Year-end or quarter-end date, fiscal year convention
  • Auditor: External auditor name and audit opinion type (unqualified, qualified, adverse, disclaimer)
  • Auditor Change: Any change in auditor is a major red flag — investigate immediately
  • Filing Date: Note if filed late (extension requests = warning signal)
  • Restatements: Any prior period restatements disclosed in current filing

Phase 2: Management Discussion & Analysis (MD&A) Deep Read

The MD&A is management's own narrative explanation of results. Read it critically — management has incentives to highlight positives and minimize negatives.

2.1 Revenue Analysis

Extract and analyze from MD&A:

Revenue Breakdown Table:
Segment/Product Line    Current Period    Prior Period    YoY Change %    Commentary
─────────────────────────────────────────────────────────────────────────────────────
[Segment 1]
[Segment 2]
[Geographic: Americas]
[Geographic: EMEA]
[Geographic: APAC]
─────────────────────────────────────────────────────────────────────────────────────
Total Revenue

Key questions to answer from MD&A:

  • What does management cite as the primary growth drivers? Are these sustainable or one-time?
  • Is growth organic (volume/price) or inorganic (acquisitions)? Separate these explicitly.
  • What is the mix shift between high-margin and low-margin products/services?
  • Is management guiding to acceleration or deceleration? Are they vague or specific?
  • What is the backlog, remaining performance obligations (RPO), or deferred revenue trend?

2.2 Margin Analysis

Margin Trend Table:
Metric              Current Period    Prior Period    QoQ or YoY    Explanation from MD&A
──────────────────────────────────────────────────────────────────────────────────────
Gross Margin %
Operating Margin %
Net Margin %
Adjusted EBITDA %   (if disclosed)
FCF Margin %        (Operating CF - Capex)
──────────────────────────────────────────────────────────────────────────────────────

Red flags in margin discussion:

  • Non-GAAP exclusions growing faster than GAAP expenses (SBC, restructuring)
  • Gross margin compression attributed to "product mix" without specifics
  • COGS increases described only in absolute terms, not as % of revenue
  • Operating leverage claims not reflected in actual operating margin expansion

2.3 Liquidity and Capital Resources

From the Liquidity section of MD&A, extract:

  • Cash and equivalents position (end of period)
  • Revolving credit facility (drawn / undrawn capacity)
  • Upcoming debt maturities (12-month and 3-year look-ahead)
  • Operating cash flow generated in period
  • Free cash flow (OCF minus capex)
  • Capital allocation priorities stated by management
  • Share repurchase authorization remaining
  • Dividend policy changes

2.4 Forward-Looking Statements Analysis

Catalog management's specific guidance language:

Guidance Item         Specific Language Used        Strength (Specific/Vague/Hedged)
────────────────────────────────────────────────────────────────────────────────────
Revenue Guidance
Margin Guidance
Capex Guidance
Hiring/Headcount
Product Launch Timing
────────────────────────────────────────────────────────────────────────────────────

Flag: When management shifts from specific numerical guidance to directional language ("we expect continued growth"), this is a meaningful signal of reduced confidence.


Phase 3: Financial Statement Deep Dive

3.1 Income Statement Quality Assessment

Line Item Analysis:
                    Current    Prior    Change    Quality Note
Revenue             $          $        %
COGS                $          $        %
Gross Profit        $          %GP
R&D                 $          %Rev
S&M                 $          %Rev
G&A                 $          %Rev
Operating Income    $          %OM
Interest Expense    $          $
Tax Rate            %          %
Net Income (GAAP)   $
EPS (GAAP)          $
Non-GAAP EPS        $         Reconciliation items: ___

Quality checks:

  • Revenue recognition policy: Has it changed? Check footnotes for ASC 606 adoption details
  • Channel stuffing signals: Receivables growing faster than revenue (days sales outstanding expanding)
  • Deferred revenue: Growing (positive — customers paying ahead) or shrinking (negative)
  • Tax rate: Unusually low tax rate often driven by one-time benefits; normalize for comparison

3.2 Balance Sheet Analysis

Balance Sheet Health Check:
Asset Quality:
  Cash & Equivalents:         $___  (% of total assets)
  Accounts Receivable:        $___  DSO: ___ days  (vs. prior: ___)
  Inventory:                  $___  DIO: ___ days  (vs. prior: ___)
  Goodwill & Intangibles:     $___  (% of total assets) — impairment risk?
  Other Long-Term Assets:     $___

Liability Structure:
  Short-Term Debt:            $___  (due within 12 months)
  Long-Term Debt:             $___
  Total Debt:                 $___
  Net Debt/(Cash):            $___
  Interest Coverage Ratio:    ___x  (EBIT / Interest Expense)
  Debt/EBITDA:                ___x

Equity Quality:
  Retained Earnings Trend:    Growing / Declining / Negative
  Treasury Stock:             $___  (buyback history)
  AOCI:                       $___  (pension obligations, FX hedges)

Balance sheet red flags:

  • Goodwill > 30% of total assets without clear acquisition rationale
  • Accounts receivable growing 2x+ faster than revenue (aggressive recognition)
  • Inventory build in a demand-softening environment
  • Off-balance sheet operating lease obligations (check footnotes — ASC 842)
  • Underfunded pension liabilities buried in AOCI

3.3 Cash Flow Statement — The Truth Teller

The cash flow statement is the hardest to manipulate. Always reconcile GAAP net income to operating cash flow:

Cash Flow Quality Analysis:
Operating Cash Flow:          $___
Less: Capex:                  $___
Free Cash Flow:               $___  FCF Margin: __%
FCF Conversion Rate:          ___   (FCF / Net Income — should be >80% for quality businesses)

Working Capital Changes:
  AR Change (use/source):     $___
  Inventory Change:           $___
  AP Change:                  $___
  Deferred Revenue Change:    $___
  Net Working Capital Impact: $___

SBC Add-Back:                 $___  (% of Revenue — >5% is dilutive)
D&A Add-Back:                 $___  (% of Revenue — maintenance capex proxy)

Cash flow quality flags:

  • FCF significantly below net income (earnings quality problem)
  • Large working capital builds consuming cash (inventory/AR growth)
  • SBC >5% of revenue — real dilution not reflected in GAAP net income
  • Capex below depreciation for multiple years (under-investment in assets)

Phase 4: Risk Factors Analysis

The Risk Factors section (Item 1A in 10-K) is legally required to disclose material risks. Read for:

4.1 New vs. Removed Risk Factors

Compare current filing's risk factors against the prior period:

Risk Factor Changes:
NEW Risks Added (first appearance):
  1. [Risk title] — [Category: Business/Regulatory/Macro/Competition/Cyber/Legal]
  2. ...

REMOVED Risks (previously disclosed, now gone):
  1. [What was removed] — [Why might this be significant?]

ELEVATED Risks (same risk, but expanded disclosure or moved higher in list):
  1. [Risk title] — [What changed in the language]

Risk categories to flag as highest priority:

  • Regulatory/Legal: New government investigations, pending litigation, regulatory changes
  • Customer Concentration: Single customer >10% of revenue disclosed
  • Technology/Competition: Specific competitor named as existential threat
  • Going Concern: Doubt about ability to continue as a going concern (immediate red flag)
  • Debt Covenant: Risk of violating debt covenants disclosed
  • Cybersecurity: New material cyber incident disclosures (Item 1.05 of 8-K)

4.2 Litigation and Legal Proceedings

From footnotes and Item 3 (Legal Proceedings):

  • Active lawsuits with >$X million exposure
  • SEC or DOJ investigations
  • Tax disputes and uncertain tax positions
  • Environmental liabilities
  • Patent litigation (key for tech and pharma)

Phase 5: Accounting Policy and Footnote Analysis

Footnotes contain what management would prefer investors not to focus on.

5.1 Critical Accounting Estimates

Management must disclose "critical accounting estimates" — areas where judgment most affects reported results:

Critical Estimate Review:
  Revenue Recognition:   [Policy and any changes]
  Goodwill Impairment:   [Testing methodology, headroom]
  Stock-Based Comp:      [Fair value models, assumptions]
  Deferred Tax Assets:   [Valuation allowance — any new allowances = warning]
  Contingent Liabilities: [What is and isn't accrued]
  Pension/OPEB:          [Return assumption, discount rate]

5.2 Related Party Transactions

Any transactions between the company and executives, board members, or major shareholders. These require heightened scrutiny for potential conflicts of interest.

5.3 Segment Reporting

When companies add or remove reporting segments, investigate:

  • Was a previously disclosed segment merged with another? (Often hides underperformance)
  • Were goodwill allocations shifted between segments? (Can delay impairment)
  • New segment disclosures can reveal previously hidden growth drivers

Phase 6: Management Tone and Language Analysis

Apply natural language analysis to management's own words:

6.1 Tone Scoring

Rate management's tone across key sections:

Tone Assessment:
Section              Tone Score (1=Very Negative → 5=Very Positive)    Key Language
──────────────────────────────────────────────────────────────────────────────────
MD&A Overview
Revenue Discussion
Margin Discussion
Guidance Language
Risk Factors Framing
CEO Letter (if present)
──────────────────────────────────────────────────────────────────────────────────
Overall Tone:        [Score] / 5
Change vs. Prior:    More Positive / Same / More Negative

6.2 Hedging Language Flags

Flag specific hedging phrases that indicate uncertainty:

  • "We believe" / "we expect" / "we anticipate" — standard forward-looking
  • "We cannot assure" / "there can be no assurance" — heightened uncertainty
  • "Significant uncertainty" / "material uncertainty" — near-crisis language
  • Passive voice for negative news ("challenges were experienced") vs. active voice for positives
  • Increasing use of "headwinds," "challenging environment," "softness"

6.3 Management Credibility Tracking

Guidance Accuracy Scorecard (prior 4 quarters):
Period    Revenue Guidance    Actual    Beat/Miss    Margin Guidance    Actual    Beat/Miss
Q-4
Q-3
Q-2
Q-1
──────────────────────────────────────────────────────────────────────────────────────────
Batting Average: ___  Beat Rate: ___  Avg. Miss: ___

Phase 7: Year-over-Year Comparison (for Quarterly Reports)

For 10-Q analysis, compare against the same quarter prior year AND the sequential prior quarter:

YoY and Sequential Comparison:
Metric          Current Q    Prior Q (seq)    Prior Year Q    YoY %    QoQ %
Revenue
Gross Margin
Operating Income
EPS (GAAP)
FCF

Identify if trends are:

  • Accelerating (sequential improvement + YoY beats)
  • Stable (sequential flat, YoY in-line)
  • Decelerating (sequential miss, YoY still positive)
  • Deteriorating (both sequential and YoY misses)

Phase 8: Insider Activity (from Proxy/Form 4 Data)

Cross-reference with available SEC Form 4 data around the filing date:

  • CEO/CFO buying on the open market (strong signal — insiders rarely buy for tax purposes)
  • Bulk insider selling at prices below historical averages (negative signal)
  • New 10b5-1 plan adoptions (scheduled selling — less informative than open market)
  • Executive departures disclosed in proxy or 8-K (CFO exits especially significant)

Output Format

Executive Summary (1-2 paragraphs)

High-level verdict on the filing quality and key takeaways for investors. Address: Is this a better or worse filing than prior periods? What is the single most important thing an investor should take away?

Financial Health Dashboard

┌─────────────────────────────────────────────────────────────────┐
│  FINANCIAL REPORT ANALYSIS — [TICKER] [Filing Type] [Period]   │
├────────────────────────┬────────────────────────────────────────┤
│ Revenue Growth         │ [%] YoY   [Organic vs. Acquired]       │
│ Gross Margin           │ [%]  (vs. prior: [%]  Δ [bps])         │
│ Operating Margin       │ [%]  (vs. prior: [%]  Δ [bps])         │
│ FCF Margin             │ [%]  (FCF Conversion: [%])              │
│ Net Debt / (Cash)      │ $[M]  (Debt/EBITDA: [x])               │
│ DSO                    │ [days]  (vs. prior: [days])             │
│ Management Tone        │ [Positive / Neutral / Cautious]         │
├────────────────────────┼────────────────────────────────────────┤
│ Filing Quality Score   │ [Score] / 10                           │
│ Earnings Quality Score │ [Score] / 10                           │
│ Balance Sheet Score    │ [Score] / 10                           │
└────────────────────────┴────────────────────────────────────────┘

Key Insights (Bulleted)

Positives (Bullish Signals):

  • [Specific insight with page/section reference from filing]

Negatives (Bearish Signals):

  • [Specific insight with page/section reference from filing]

Neutral / Watch Items:

  • [Items requiring monitoring but not yet actionable]

Red Flags and Anomalies

#Red FlagSeverityWhere FoundImplication
1HIGH/MED/LOW

Accounting Quality Score

CriterionScore (0–3)Notes
Revenue recognition clarity
Non-GAAP reconciliation quality
FCF conversion rate
Working capital trends
Footnote transparency
Auditor opinion
Related party transactions
Total/21

Score guide: 18–21 = High quality | 12–17 = Average | <12 = Scrutinize further


Input Formats Accepted

# Paste raw text from 10-K or 10-Q sections
/financial-report-analyst [paste filing text]

# Analyze specific section of a recent filing
/financial-report-analyst AAPL 10-K --section md&a
/financial-report-analyst MSFT 10-Q --section risk-factors
/financial-report-analyst NVDA 10-K --section cash-flow

# Full report analysis (all sections)
/financial-report-analyst GOOGL 10-K --full

# Compare current vs prior filing
/financial-report-analyst META 10-K --compare-prior

# Quick earnings press release analysis
/financial-report-analyst AMZN --press-release [paste text]

# Proxy statement analysis
/financial-report-analyst TSLA --proxy

Thesis Invalidation

After delivering the analysis signal, specify what would reverse it:

If signal is BULLISH — thesis breaks if:

  • auditor issues going concern warning OR material restatement announced
  • [One or two more triggers drawn from this analysis's own drivers, each with a threshold]

If signal is BEARISH — thesis breaks if:

  • new auditor clears all concerns AND restatement resolved favorably
  • [One or two more triggers drawn from this analysis's own drivers, each with a threshold]

Re-run this analysis when:

  • Next earnings release
  • Price moves ±15% from current level
  • 60 days have elapsed
  • Material news event (acquisition, leadership change, regulatory decision)

Standard Signal Output

All analysis concludes with this standardized block:

In the output, the box below is followed immediately by this machine-readable footer — the same values as the box, filled in (no placeholders, no comments), so tools can read the result without parsing the box:

{"investskill": "1", "skill": "financial-report-analyst", "ticker": "<TICKER, or null for a portfolio / market-wide view>", "as_of": "<YYYY-MM-DD from Data & Sources>", "signal": "<BULLISH | NEUTRAL | BEARISH>", "confidence": "<HIGH | MEDIUM | LOW>", "horizon": "<SHORT-TERM | MEDIUM-TERM | LONG-TERM>", "score": 0.0, "action": "<BUY | HOLD | SELL>", "conviction": "<STRONG | MODERATE | WEAK>", "retrieval": "<pasted by user | web/tool retrieval | model memory>", "sources": ["<each primary source, dated>"]}
╔══════════════════════════════════════════════╗
║              INVESTMENT SIGNAL               ║
╠══════════════════════════════════════════════╣
║ Signal:      BULLISH / NEUTRAL / BEARISH     ║
║ Confidence:  HIGH / MEDIUM / LOW             ║
║ Horizon:     SHORT / MEDIUM / LONG-TERM      ║
║ Score:       X.X / 10                        ║
╠══════════════════════════════════════════════╣
║ Action:      BUY / HOLD / SELL               ║
║ Conviction:  STRONG / MODERATE / WEAK        ║
╚══════════════════════════════════════════════╝

Score Guide: 8.0–10.0 Strongly Bullish | 6.0–7.9 Moderately Bullish | 4.0–5.9 Neutral | 2.0–3.9 Moderately Bearish | 0.0–1.9 Strongly Bearish Confidence: HIGH (strong data, clear signals) | MEDIUM (mixed signals) | LOW (limited data, conflicting signals) Horizon: SHORT-TERM (1 week–3 months) | MEDIUM-TERM (3 months–1 year) | LONG-TERM (1+ years)

Disclaimer: Educational analysis only. Not financial advice.

Forensic Accounting — Are the Reported Earnings Real?

⚠️ Data Verification — Do This Before Any Analysis

Before running any analysis, always retrieve the latest market data for the ticker:

  1. Fetch current price — use web search or ask the user for the live price, 52-week range, and market cap. Never assume a price from training data. (Market cap feeds Altman's X4; everything else in this skill comes from the filings.)
  2. Pull the filings, not a data feed — this skill needs at least two, preferably three to five, consecutive fiscal years of as-filed statements, plus their notes:
    • 10-K Item 8: income statement, balance sheet, cash-flow statement (for every index below you need year t and year t−1).
    • Revenue note (ASC 606): disaggregation, contract assets / unbilled receivables, deferred revenue, remaining performance obligations, any bill-and-hold or principal-vs-agent judgment.
    • Receivables / allowance note, inventory note (reserves, LIFO/FIFO changes), PP&E and intangibles notes (capitalised software, useful-life changes), capitalised contract costs (ASC 340-40), related-party note, stock-compensation note, commitments and contingencies.
    • Auditor's report (opinion, critical audit matters, going-concern paragraph, tenure), Item 9A (internal control over financial reporting — material weaknesses), and 8-K Item 4.01 (auditor change) / Item 4.02 (non-reliance — restatement) for the last five years.
    • Earnings releases for the non-GAAP reconciliation (8-K Exhibit 99.1), and Form 4s for insider sales around results. Use as-filed figures; when a later filing restated a prior year, note both and compute the indices on the restated basis (the restatement itself is a finding).
  3. State your data source — fill in the Data & Sources header (next section) so the origin, as-of date, retrieval path, and confidence of every figure are explicit at the top of the output.
  4. Flag stale data explicitly — if live data is unavailable, display this warning before proceeding:

⚠️ Live data unavailable. The following analysis uses training-data estimates which may be significantly out of date. Verify all prices and metrics before making any decisions.

Never silently substitute training-data estimates for current prices. When in doubt, ask the user to paste the latest quote.


📋 Data & Sources Header — Open Every Output With It

The first thing in the output is this provenance block, filled in — never left as placeholders. It is the standard documented on the Data & Accuracy page and the first thing result-validator looks for:

Data & Sources
  As of:      <date the figures represent, e.g. 2026-06-30>
  Source:     <primary docs — SEC EDGAR 10-K/10-Q, company IR, FRED, exchange data …>
  Retrieval:  <pasted by user | web/tool retrieval | model memory>
  Confidence: <HIGH | MEDIUM | LOW>
  • Retrieval: model memory must be paired with Confidence: LOW — memory is a placeholder until confirmed against a primary source.
  • Mixed sources: list each with its own as-of date rather than blending them.
  • Sources disagree on the same figure (a closing price, a deal term): prefer the primary source (SEC filing, company release, exchange data) over an aggregator, then the most recent dated one. Show both values with their sources and dates, say which you used and why, and never average them.
  • Data the user pasted is reported as pasted by user; do not upgrade its confidence beyond what the user's own source supports.

🌐 Output Language — --lang zh-TW

English is the default. With --lang zh-TW — or when the user writes in Traditional Chinese and asks for a Chinese answer — write the whole analysis in Traditional Chinese (繁體中文, Taiwan usage), keeping it machine-checkable:

  • Translate every heading, narrative paragraph, table header and bullet. Give each metric's English name in brackets the first time it appears — 自由現金流(Free Cash Flow, FCF) — so the reader can match it to the glossary and to English analyses.
  • Keep in English tickers, figures and units, SEC form and item names (10-K Item 7, Form 4), the four Data & Sources field labels (As of, Source, Retrieval, Confidence — the Retrieval value may be written in Chinese, e.g. 使用者提供), and the Investment Signal box and JSON footer exactly as specified — field names and the BULLISH / NEUTRAL / BEARISH, HIGH / MEDIUM / LOW, BUY / HOLD / SELL, STRONG / MODERATE / WEAK values. Tools parse them.
  • Title the invalidation section 論點失效條件(Thesis Invalidation), and close with the disclaimer in Chinese: 免責聲明: 僅供教育用途,非投資建議。
  • Use Taiwan financial terms: 殖利率、本益比、營業利益率、自由現金流、聯準會、選擇權、買權/賣權、除息日、預扣稅.

Overview

Every valuation in this catalog takes reported earnings as an input. This skill asks the question that comes first: how much of that earnings number is cash, how much is judgment, and do the patterns in the statements look like the patterns that preceded past manipulations and failures? It runs the academic screens that have a track record (Beneish's M-Score, Altman's Z, Sloan's accruals anomaly), then the practitioner checks that catch what the models miss — revenue-recognition stretch, cost capitalisation, the widening gap between "adjusted" and GAAP earnings, stock compensation treated as free, and the governance tells (auditor changes, material weaknesses, restatements, related parties).

This is a screen for risk, not an accusation. A high reading means "the statements have the shape of trouble and deserve a skeptical read," not "this company commits fraud." Most companies that trip one flag have an innocent explanation (an acquisition distorts DSRI; a fast grower has high SGI by definition). The skill's job is to compute the numbers exactly, explain each flag, and say which explanation the filings support.

Where it sits. financial-report-analyst reads the filing and gives an Accounting Quality Score; 10k-digest summarises it; bear-case argues the short thesis and mentions Beneish without computing it. This skill does the arithmetic and returns a Manipulation Risk Score (0–10, higher = more risk). Feed a high score into bear-case as evidence, check any surprising figure with fact-check, and read insider selling around the flagged periods with insider-trading.


1. Inputs

InputRequiredDefault if unstated
Ticker✅—
Fiscal years to coverrecommendedlatest 3 fiscal years (2 is the minimum for any index; 5 for trend checks)
Filings pasted by the user (10-K, 10-Q, earnings release)optionalretrieve from SEC EDGAR; say which accession numbers were used
Sector override (--sector bank / insurer / reit)optionalinferred from SIC code; changes which models apply (§2, Phase 7)
Peer tickers for DSO / accrual comparisonoptional2–3 closest peers named in the 10-K or by competitor-analysis
Specific concern ("short report says channel stuffing")optionalnone — run the full screen

Quarterly mode. If the user asks about a recent quarter, compute DSO/DIO/DPO, unbilled receivables and the non-GAAP gap on trailing-twelve-month figures and the last eight quarters; run Beneish and Altman on fiscal years only (the coefficients were estimated on annual data).


2. Framework

Phase 1 — Beneish M-Score (8-variable model)

Compute each index from year t (latest) and year t−1. Show every input in a table so the reader can check the arithmetic.

IndexFormulaWhat a high value suggestsNon-manipulator meanManipulator mean
DSRI Days' Sales in Receivables Index(Receivables_t / Sales_t) ÷ (Receivables_t−1 / Sales_t−1)Receivables growing faster than sales — revenue booked before cash1.0311.465
GMI Gross Margin IndexGM_t−1 ÷ GM_t, where GM = (Sales − COGS) / SalesDeteriorating margins — pressure to manage earnings1.0141.193
AQI Asset Quality Index[1 − (Current assets_t + Net PP&E_t + Securities_t) / Total assets_t] ÷ [same for t−1]More costs parked in soft, non-current assets (capitalisation)1.0391.254
SGI Sales Growth IndexSales_t ÷ Sales_t−1Growth firms face pressure to sustain it; not manipulation by itself1.1341.607
DEPI Depreciation Index[Dep_t−1 / (Dep_t−1 + Net PP&E_t−1)] ÷ [Dep_t / (Dep_t + Net PP&E_t)]Slower depreciation — useful lives lengthened1.0011.077
SGAI SG&A Index(SG&A_t / Sales_t) ÷ (SG&A_t−1 / Sales_t−1)SG&A rising relative to sales — weaker efficiency1.0541.041
LVGI Leverage Index[(Current liabilities_t + Total long-term debt_t) / Total assets_t] ÷ [same for t−1]Rising leverage — debt-covenant incentives1.0371.111
TATA Total Accruals to Total Assets(Income from continuing operations_t − Cash from operations_t) / Total assets_tEarnings not backed by cash0.0180.031

Reference means are from Beneish (1999), "The Detection of Earnings Manipulation," Financial Analysts Journal.

The model:

M = −4.84 + 0.920·DSRI + 0.528·GMI + 0.404·AQI + 0.892·SGI
          + 0.115·DEPI − 0.172·SGAI + 4.679·TATA − 0.327·LVGI
M-ScoreReading
> −1.78Likely manipulator zone (Beneish's original cutoff)
−2.22 to −1.78Grey zone — the widely used, more conservative −2.22 cutoff (originally paired with the 5-variable model) flags it; the original does not
< −2.22Unlikely manipulator

Report the M-Score against both cutoffs and name the one or two indices that drive it (largest coefficient × deviation from the non-manipulator mean). A score above the threshold driven only by SGI (a genuine fast grower) is weaker evidence than one driven by DSRI + TATA + AQI.

Computation rules.

  • Securities in AQI = long-term investments (not cash). If the balance sheet shows none, use 0 and say so.
  • Depreciation excludes amortisation of acquired intangibles where the filing separates them; Net PP&E excludes operating-lease right-of-use assets (ASC 842 inflates AQI otherwise — note it if the adoption year is in the window).
  • A major acquisition in year t distorts DSRI, AQI and SGI. Flag it and, if the target's pro-forma figures are disclosed, recompute on a pro-forma basis.
  • The model was estimated on non-financial firms; do not run it on banks, insurers or REITs (Phase 7).

Phase 2 — Altman Z (distress, not fraud — but distress is a motive)

Choose the version by the company's profile, state which you used and why:

VersionUse forFormulaSafeGreyDistress
Z (1968)Public manufacturers / asset-heavy industrials1.2·X1 + 1.4·X2 + 3.3·X3 + 0.6·X4 + 1.0·X5> 2.991.81 – 2.99< 1.81
Z' (1983)Private manufacturers (no market cap)0.717·X1 + 0.847·X2 + 3.107·X3 + 0.420·X4′ + 0.998·X5> 2.901.23 – 2.90< 1.23
Z'' (1995)Non-manufacturers (services, software, retail) and emerging-market issuers6.56·X1 + 3.26·X2 + 6.72·X3 + 1.05·X4′> 2.601.10 – 2.60< 1.10

Where X1 = working capital / total assets; X2 = retained earnings / total assets; X3 = EBIT / total assets; X4 = market value of equity / total liabilities; X4′ = book equity / total liabilities; X5 = sales / total assets. Z'' drops X5 because asset turnover varies too much across non-manufacturing industries. For emerging-market issuers, Altman's EM score adds a constant of 3.25 to Z'' (zones > 5.85 / 4.35–5.85 / < 4.35).

Caveats to print. Large buybacks drive retained earnings (X2) negative for healthy companies — say so rather than calling them distressed. Asset-light software firms with big deferred revenue look worse on X1 than they are. Never apply Z to banks, insurers, or REITs.

Phase 3 — Accruals and cash conversion

Sloan accruals ratio — balance-sheet version (Sloan 1996):

Accruals_BS = (ΔCurrent assets − ΔCash) − (ΔCurrent liabilities − ΔShort-term debt − ΔTaxes payable) − Depreciation & amortisation
Accruals ratio_BS = Accruals_BS / Average total assets

Cash-flow version (cleaner; unaffected by acquisitions and reclassifications):

Accruals_CF = Net income − (Cash from operations + Cash from investing)
Accruals ratio_CF = Accruals_CF / Average net operating assets
   (NOA = (Total assets − Cash & investments) − (Total liabilities − Total debt))
Accruals ratio (either version)Reading
between −10% and +10%Normal
+10% to +20%Elevated — earnings running ahead of cash; explain it
> +20%Red flag — Sloan's top accrual decile historically underperformed
< −10%Conservative accounting or heavy investment (check CFI) — usually benign

Cash conversion. OCF / Net income for each of the last 3–5 years, and the cumulative ratio (Σ OCF / Σ NI) across the window. Cumulative ≥ 1.0 is healthy for a mature company; below 0.8 cumulatively is a flag. One bad year is noise; a declining trend while net income rises is the classic divergence. Also check where OCF comes from: a one-off boost from stretching payables, selling receivables (factoring), or supplier-finance programmes is not durable — look for it in the working-capital lines and the supplier-finance disclosure (ASU 2022-04).

SBC-adjusted free cash flow.

FCF          = OCF − Capex (including capitalised software)
SBC-adj. FCF = FCF − Stock-based compensation
SBC % revenue = SBC / Revenue
SBC % of revenueReading
< 5%Normal for most sectors
5% – 15%Material — value FCF only on the SBC-adjusted line
> 15%Severe — reported FCF overstates owner earnings; check net share count growth

Report net dilution (diluted share count change, after buybacks) alongside it: buybacks that only offset SBC are a compensation expense paid in cash.

Phase 4 — Revenue recognition and working capital

Working-capital days, 3–5 years, vs. 2–3 peers:

DSO = Accounts receivable / Revenue × 365
DIO = Inventory / COGS × 365
DPO = Accounts payable / COGS × 365
CCC = DSO + DIO − DPO

Flag any of these rising more than 15% year on year without a disclosed reason (mix shift, new large customer with longer terms, acquisition). DSO rising while revenue accelerates is the core channel-stuffing pattern; DIO rising with flat sales signals obsolescence the reserve may not cover; DPO rising sharply flatters OCF temporarily.

Revenue-recognition red flags — check each, mark ✅ none found / ⚠️ present / ❓ not disclosed:

FlagWhere to lookWhat it looks like
Unbilled receivables / contract assets growing faster than revenueRevenue note (ASC 606 contract balances)Revenue recognised on progress or milestones before the customer can be billed — judgment-heavy
Channel stuffingMD&A, receivables, distributor disclosures, quarterly revenue shapeDSO up, quarter-end revenue spike, distributor inventory rising, extended payment terms or rights of return, next-quarter weakness
Bill-and-holdRevenue policy noteRevenue on goods the company still holds; ASC 606 allows it only under strict criteria — any mention needs scrutiny
Principal vs. agentRevenue policyGross revenue booked on pass-through transactions inflates growth (not profit)
Changes in estimatesPolicy note, CAMsShift in percentage-of-completion inputs, variable consideration, returns reserves, useful lives
Deferred revenue falling while revenue risesBalance sheetFor subscription businesses, a leading indicator that future revenue is weakening or terms changed
Allowance for doubtful accounts falling as % of receivablesReceivables noteReserve release flattering earnings
Quarter-end concentration / large "other" revenue10-Q segment dataRevenue that does not fit the segment story

Phase 5 — Capitalised costs and the GAAP vs. non-GAAP gap

Capitalisation. For each category, report the amount capitalised in the year, as % of the related expense, and the trend:

CategoryStandardFlag when
Internal-use / for-sale softwareASC 350-40 / ASC 985-20Capitalised share of total software spend rising; amortisation period lengthened
Development costs (R&D)US GAAP expenses most R&D; IFRS (IAS 38) capitalises developmentIFRS filer whose capitalised development grows faster than revenue
Contract acquisition costs (commissions)ASC 340-40Asset growing faster than revenue; amortisation period longer than the customer life
Capitalised interest, customer-acquisition or "pre-opening" costsVariousAny new capitalisation policy in the window

Capitalising costs moves them from the income statement and from OCF into investing cash flow — so check capex including capitalised software, not just PP&E capex.

Non-GAAP vs. GAAP gap. For each of the last 3–5 years (and 8 quarters):

Gap % = (Non-GAAP net income − GAAP net income) / |GAAP net income|

Itemise the adjustments. Flag: a gap widening for three consecutive periods; "one-time" restructuring charges that recur in most years; SBC excluded and buybacks presented as returning capital; acquisition-related amortisation excluded while acquisitions are the growth strategy; adjustments the company added this year.

Phase 6 — Governance, audit, and related parties

CheckSourceRisk reading
Auditor change in last 5 years8-K Item 4.01Resignation (vs. dismissal) and any reported disagreement or reportable event is a major flag; a move from a large audit firm to a much smaller one is a flag
Critical audit mattersAuditor's reportList them; CAMs on revenue recognition, inventory valuation, or capitalisation overlap with Phases 4–5 — read them together
Going-concern paragraphAuditor's report, MD&ASubstantial doubt → hard floor (below)
Material weakness in ICFRItem 9AAny MW in the last 3 years; worse if it relates to revenue or the financial close, or remediation keeps slipping
Restatement / non-reliance8-K Item 4.02, "as restated" labels"Big R" restatement (prior statements no longer reliable) is a major flag; "little r" revision less so — say which
Late filingForm 12b-25 (NT 10-K / NT 10-Q)Late filing plus auditor issues is the pattern that precedes most restatements
Related-party transactionsRelated-party note, proxySales to, purchases from, leases with, or loans to entities controlled by insiders; size vs. revenue or SG&A
CFO / CAO turnover8-K Item 5.02More than one finance-chief change in 3 years
Insider selling around resultsForm 4Clustered sales by the CEO/CFO shortly before a weak print or a restatement — hand to insider-trading for the detail
SEC inquiries, subpoenas, short-seller reportsContingencies note, 8-Ks, newsNote the allegation and whether the filings support or rebut it — do not adjudicate

Phase 7 — Sector exceptions (financials, insurers, REITs)

The standard models assume an operating company. For these sectors, do not compute the listed models — say they do not apply and use the substitutes:

SectorDo not applyUse instead
Banks / lendersBeneish (excluded from its estimation sample), Altman Z / Z' / Z'', DSO/DIO, cash conversion (OCF is dominated by loan and deposit flows)Provision for credit losses vs. net charge-offs (reserve releases flattering EPS), allowance / loans trend, Level 3 assets as % of equity, held-to-maturity unrealised losses (AOCI), non-GAAP "core" adjustments
InsurersBeneish, Altman, DSO/DIO, OCF/NIPrior-year reserve development (repeated favourable releases, then a large strengthening), reserve-to-premium trends, reinsurance with related or offshore entities, Level 3 investments
REITsAltman X2 (REITs distribute nearly all income), EPS-based accrualsFFO / AFFO reconciliation (what is added back), straight-line rent and capitalised leasing costs as % of AFFO, capitalised interest on development, related-party management fees

Governance checks (Phase 6) apply to every sector. For these companies the score is built from Phase 6 plus the substitutes; renormalise the weights (Phase 8) and say so.

Phase 8 — Manipulation Risk Score (0–10, higher = more risk)

Score each component 0 (clean) to 10 (severe) on the anchors, then weight:

ComponentWeight0 looks like5 looks like10 looks like
Beneish M-Score20%M < −2.50, no index above the manipulator mean−2.22 < M < −1.78, or M > −1.78 driven only by SGIM > −1.78 driven by DSRI, TATA, or AQI
Accruals & cash conversion20%Accruals ratio within ±10%; cumulative OCF/NI ≥ 1.0Accruals 10–20%, or OCF/NI 0.8–1.0Accruals > 20% and OCF/NI < 0.8, or OCF propped up by factoring / supplier finance
Revenue recognition & working capital20%Stable DSO/DIO vs. peers; no red flagsOne flag with a plausible disclosed reasonDSO and unbilled receivables rising with revenue acceleration; bill-and-hold or rising returns rights
Capitalisation, non-GAAP & SBC15%Stable capitalisation; gap < 10%; SBC < 5% of revenueGap widening, or SBC 5–15%New/expanding capitalisation policy, gap widening 3 periods, SBC > 15%
Governance & audit15%Long-tenured auditor, no MW, no restatement, minor related partiesOne CAM in a sensitive area or a remediated MWAuditor resignation, unremediated MW, Big-R restatement, material related-party revenue
Distress motive (Altman)10%Safe zoneGrey zoneDistress zone
Manipulation Risk Score (MRS) = Σ weight × component score
Box Score                     = 10 − MRS

Hard floors on MRS (state any that fired):

  • Going-concern paragraph, auditor resignation with a disagreement, or a Big-R restatement in the last 3 years → MRS ≥ 7.0
  • Unremediated material weakness, late 10-K filing, or an open SEC enforcement inquiry → MRS ≥ 6.0

Missing data. If a component cannot be computed (one year of data only, sector exception), drop it and renormalise the remaining weights to 100%, list what was dropped, and cap Confidence at MEDIUM. Never score an uncomputed component as 0.

MRSReadingBox ScoreSignal
0.0 – 4.0Low risk — earnings appear backed by cash and conservative policy6.0 – 10.0BULLISH
4.1 – 6.0Moderate — specific items need monitoring4.0 – 5.9NEUTRAL
6.1 – 10.0High — treat reported earnings as unreliable until explained0.0 – 3.9BEARISH

3. Output Format

  1. Data & Sources header (above), listing the filings (form, fiscal year, accession number) used
  2. Executive Summary — three sentences: the MRS and what drives it, the single most important flag, what to do (monitor / dig further / avoid until explained)
  3. Beneish table — eight indices with inputs, t and t−1 values, the M-Score vs. −1.78 and −2.22, and the driving indices
  4. Altman table — version used and why, X1–X5, Z, zone, and any caveat (buybacks, deferred revenue)
  5. Accruals & cash — Sloan ratio (both versions), OCF/NI by year and cumulative, FCF vs. SBC-adjusted FCF, SBC % revenue, net dilution
  6. Revenue & working capital — DSO/DIO/DPO/CCC trend vs. peers; red-flag checklist (✅ / ⚠️ / ❓)
  7. Capitalisation & non-GAAP — capitalised amounts and trend; GAAP vs. non-GAAP bridge by year
  8. Governance & audit — auditor, CAMs, going concern, MW, restatements, related parties, insider sales
  9. Manipulation Risk Score — component table, hard floors fired, MRS, box Score = 10 − MRS
  10. Thesis Invalidation — what would reverse the call (below)
  11. Investment Signal Block — the standard box (below)

Example

User: Run a forensic accounting check on ZEPH (fictional ticker, FY2024–FY2025 10-Ks pasted)

The assistant fills the Data & Sources header (two 10-Ks, pasted by user, HIGH),
then computes the Beneish indices. Hypothetical inputs: revenue grew from $800M to
$1,000M (SGI 1.25) while receivables grew from $120M to $200M, so DSRI =
(200/1,000) / (120/800) = 1.33 — well above the non-manipulator mean of 1.03.
Net income of $90M against OCF of $40M on $1,600M of assets gives TATA = 0.031,
at the manipulator mean. With GMI 1.02, AQI 1.10, DEPI 1.05, SGAI 0.95 and
LVGI 1.00, M = −4.84 + 1.227 + 0.539 + 0.444 + 1.115 + 0.121 − 0.163 + 0.146
− 0.327 = −1.74: above both cutoffs, driven by DSRI and SGI.

Z'' (ZEPH is a software company) sits in the safe zone, so there is no distress
motive. The cash-flow accruals ratio is +14% and cumulative OCF/NI over three years
is 0.62. The revenue note shows unbilled receivables up 140% against revenue up
25%, and capitalised commissions growing faster than revenue. The auditor is
unchanged and there is no material weakness.

Components: Beneish 8, accruals 8, revenue 8, capitalisation 6, governance 2,
distress 1 → MRS 6.1, box Score 3.9, BEARISH, MEDIUM confidence (two years only;
peer DSO not supplied). Next steps: `fact-check` the receivables figures,
`earnings-call-analysis` for management's explanation of unbilled growth,
`bear-case` to build the full short thesis.

Notes

  • Screens, not verdicts. Beneish's model catches a meaningful share of later-revealed manipulators but also flags many innocent firms; Altman's Z predicts distress, not fraud. Use the language "consistent with" and "warrants scrutiny," never "the company is manipulating earnings."
  • Never invent figures. Every index needs two years of exact line items; if they were not retrieved or pasted, mark the index "not computed" rather than estimating it from memory.
  • Show the arithmetic. Print the inputs to every ratio. A forensic report the reader cannot re-check is worthless — and fact-check will re-check it.
  • Context beats thresholds. An acquisition, a change in fiscal year, ASC 606 / ASC 842 adoption, or a business-model shift (hardware to subscription) breaks year-over-year indices. Identify the event and, where possible, compute on a like-for-like basis.
  • Fast growers trip SGI and DSRI honestly. Compare DSO with peers and with the company's own history before calling it stuffing.
  • Short-seller reports are allegations; test their specific claims against the filings and report which are supported, contradicted, or unverifiable.
  • Run before: 10k-digest or financial-report-analyst (to orient in the filing). Run after: bear-case (to argue the risk), fact-check (to verify any figure that drives the score), insider-trading (for sales around flagged periods), credit-check (if the Altman reading is grey or distressed).

Thesis Invalidation

After delivering the analysis signal, specify what would reverse it:

If signal is BULLISH (low manipulation risk) — thesis breaks if:

  • DSO or unbilled receivables rise more than 15% year on year without a disclosed reason in the next 10-Q, or cumulative OCF/NI falls below 0.8
  • The auditor resigns, a material weakness is disclosed, an 8-K Item 4.02 non-reliance notice is filed, or a 10-K is filed late (NT 10-K)
  • The non-GAAP gap widens for a third consecutive period or a new capitalisation policy appears
  • A credible short-seller report or SEC inquiry makes specific claims the filings cannot rebut

If signal is BEARISH (high manipulation risk) — thesis breaks if:

  • The flagged receivables or contract assets convert to cash in the next one to two quarters (OCF catches up with net income)
  • The driver of the M-Score is explained by an identified event (acquisition, ASC 606 transition) and the like-for-like recomputation falls below −2.22
  • A material weakness is remediated with a clean ICFR opinion, or a new auditor issues a clean opinion without restating
  • The company narrows its non-GAAP adjustments or begins expensing previously capitalised costs

Re-run this analysis when:

  • Next 10-K is filed (the models are annual) — and each 10-Q for the working-capital and non-GAAP checks
  • Any 8-K Item 4.01 (auditor change), 4.02 (non-reliance), or 5.02 (CFO departure)
  • An NT 10-K / NT 10-Q late-filing notice, or a short-seller report is published
  • Price moves ±15% from current level
  • Material news event (acquisition, leadership change, regulatory decision)

Standard Signal Output

This skill measures reporting risk, not valuation or direction, so state the mapping: Signal: BULLISH means the reported earnings look trustworthy enough to use as an input for valuation; NEUTRAL means specific items need monitoring before relying on them; BEARISH means treat the reported earnings as unreliable until the flags are explained. Action: SELL here means "do not underwrite a position on these numbers" — read whether the stock is cheap from stock-valuation, not from this block.

All analysis concludes with this standardized block:

In the output, the box below is followed immediately by this machine-readable footer — the same values as the box, filled in (no placeholders, no comments), so tools can read the result without parsing the box:

{"investskill": "1", "skill": "forensic-accounting", "ticker": "<TICKER, or null for a portfolio / market-wide view>", "as_of": "<YYYY-MM-DD from Data & Sources>", "signal": "<BULLISH | NEUTRAL | BEARISH>", "confidence": "<HIGH | MEDIUM | LOW>", "horizon": "<SHORT-TERM | MEDIUM-TERM | LONG-TERM>", "score": 0.0, "action": "<BUY | HOLD | SELL>", "conviction": "<STRONG | MODERATE | WEAK>", "retrieval": "<pasted by user | web/tool retrieval | model memory>", "sources": ["<each primary source, dated>"]}
╔══════════════════════════════════════════════╗
║              INVESTMENT SIGNAL               ║
╠══════════════════════════════════════════════╣
║ Signal:      BULLISH / NEUTRAL / BEARISH     ║
║ Confidence:  HIGH / MEDIUM / LOW             ║
║ Horizon:     SHORT / MEDIUM / LONG-TERM      ║
║ Score:       X.X / 10                        ║
╠══════════════════════════════════════════════╣
║ Action:      BUY / HOLD / SELL               ║
║ Conviction:  STRONG / MODERATE / WEAK        ║
╚══════════════════════════════════════════════╝

Score Guide: 8.0–10.0 Strongly Bullish | 6.0–7.9 Moderately Bullish | 4.0–5.9 Neutral | 2.0–3.9 Moderately Bearish | 0.0–1.9 Strongly Bearish Confidence: HIGH (strong data, clear signals) | MEDIUM (mixed signals) | LOW (limited data, conflicting signals) Horizon: SHORT-TERM (1 week–3 months) | MEDIUM-TERM (3 months–1 year) | LONG-TERM (1+ years)

Note: The box Score is 10 − Manipulation Risk Score, so higher manipulation risk produces a lower Score and a more bearish signal (MRS 7.0 → Score 3.0 → BEARISH); always print the MRS itself next to the box. Confidence is at most MEDIUM when fewer than three fiscal years were analysed or any component was dropped, and LOW when the figures came from model memory.

Disclaimer: Educational analysis only. Not financial advice.

Full Report — Comprehensive Stock Analysis Orchestrator

Orchestrate InvestSkill analysis modules on one ticker at your chosen depth, synthesize findings into a unified investment thesis with composite scoring, and produce a standalone professional HTML report with interactive Chart.js visualizations.

Data Verification

Open the report with a Data & Sources header (see Data & Accuracy) so provenance is explicit:

Data & Sources
  As of:      <date the figures represent>
  Source:     <primary docs — SEC EDGAR 10-K/10-Q, company IR, FRED, …>
  Retrieval:  <pasted by user | web/tool retrieval | model memory>
  Confidence: <HIGH | MEDIUM | LOW>

Before rendering the final report, verify:

  • The Data & Sources header is present and the Retrieval/Confidence are honest (flag "model memory" as LOW)
  • One snapshot: every module ran from the same data, gathered once in Workflow step 2 — one price, one share count and one as-of date for the whole report. A module that quotes a different price for the ticker than the cover page is a defect; fix it at the source, do not average
  • All executed modules returned scores (not blank)
  • Scores are within 0–10 range
  • Weighted composite arithmetic is correct (weights sum to 100%)
  • Missing modules for quick/standard depth are flagged as "not assessed" and the re-weighted component weights are shown
  • No data older than 90 days used without explicit warning

Then run fact-check on the composite (every figure the modules produced, verified against the filing and market data, with citations), and hand result-validator two separate things: the corrected composite as the analysis to validate (so its "Original Score" is the investment score, not the Verification Score), and the fact-check ledger and Verification Score as evidence for its Data Quality dimension. Include both scores in the report footer, labelled — a composite thesis should never ship without this verification-and-validation pass.


🌐 Output Language — --lang zh-TW

English is the default. With --lang zh-TW — or when the user writes in Traditional Chinese and asks for a Chinese answer — write the whole analysis in Traditional Chinese (繁體中文, Taiwan usage), keeping it machine-checkable:

  • Translate every heading, narrative paragraph, table header and bullet. Give each metric's English name in brackets the first time it appears — 自由現金流(Free Cash Flow, FCF) — so the reader can match it to the glossary and to English analyses.
  • Keep in English tickers, figures and units, SEC form and item names (10-K Item 7, Form 4), the four Data & Sources field labels (As of, Source, Retrieval, Confidence — the Retrieval value may be written in Chinese, e.g. 使用者提供), and the Investment Signal box and JSON footer exactly as specified — field names and the BULLISH / NEUTRAL / BEARISH, HIGH / MEDIUM / LOW, BUY / HOLD / SELL, STRONG / MODERATE / WEAK values. Tools parse them.
  • Title the invalidation section 論點失效條件(Thesis Invalidation), and close with the disclaimer in Chinese: 免責聲明: 僅供教育用途,非投資建議。
  • Use Taiwan financial terms: 殖利率、本益比、營業利益率、自由現金流、聯準會、選擇權、買權/賣權、除息日、預扣稅.

Depth Flags

Select how many modules to run with --depth, and drop any with --skip:

FlagModulesBest For
--depth quick5 modules — one per scored componentRapid screen, time-constrained due diligence
--depth standard10 modulesBalanced research before initiating a position
--depth comprehensiveEvery framework that applies to one stock — 20 analysis modules + 5 that build the plan, the charts, the verification and the saved thesis (default)Full pre-commitment due diligence
--skip a,b,…Removes the named modules at any depth (e.g. --skip options-analysis,short-interest for a stock without listed options). Skipped modules are listed as "skipped by request" and their component is re-weighted (below)Cost control, or a module that does not apply

Module Sets by Depth

Quick (5 modules — every scored component covered once)

#ModuleComponentFocus
1stock-evalBusiness QualityCompany overview, financial strength, relative valuation
2stock-valuationValuationDCF intrinsic value (bear/base/bull), P/E, EV/EBITDA, peer multiples
3insider-tradingMarket SignalsSEC Form 4 patterns, net insider sentiment
4technical-analysisTechnical SetupMA, RSI, MACD, volume, support/resistance
5bear-caseRisk ProfileShort-seller red-team, Thesis-Killers, downside target

Standard (10 modules — Quick + 5 more)

#ModuleComponentFocus
6competitor-analysisBusiness QualityMoat, market share, Porter's Five Forces
7financial-report-analystBusiness Quality10-K/10-Q statement quality, red flags, risk factors
8earnings-call-analysisMarket SignalsManagement tone, guidance quality, key themes
9institutional-ownershipMarket Signals13F holdings, smart-money flows
10catalyst-calendarRisk ProfileDated events in the next 90 days and their expected impact

Comprehensive (every framework that applies to one stock — Standard + 15 more)

#ModuleComponentFocus
1110k-digest(input)Structured digest of the latest 10-K; financial-report-analyst reads it and the report's appendix cites it
12industry-mapBusiness QualityPosition in the value chain, chokepoints, margin pools
13dividend-analysisBusiness QualityCapital allocation: dividends, buybacks, FCF deployment, M&A, debt management
14earnings-previewMarket SignalsConsensus vs. whisper, implied move, three-scenario grid — only if the next report is within 45 days; otherwise listed as "not applicable this quarter"
15sector-analysisTechnical SetupSector rotation, relative strength vs. benchmark
16short-interestRisk ProfileShort ratio, days to cover, squeeze risk
17options-analysisRisk ProfileIV and IV rank, put/call, what the options market prices in
18economics-analysisRisk ProfileMacro regime and rate sensitivity
19forensic-accountingRisk ProfileBeneish, Altman, accruals, cash conversion — earnings quality
20credit-checkRisk ProfileMaturity wall, coverage, refinancing cost — skipped automatically for banks and insurers
21position-ladder(plan)The Entry Strategy: rungs, share-count floor and ceiling, thesis-break exit
22chart-master(charts)The report's MA / RSI / MACD / volume / scorecard charts
23fact-check(verification)Every figure checked against the filing and market data, with citations
24result-validator(verification)Confidence across data quality, methodology and signal consistency
25thesis-tracker(saved thesis)Writes output/thesis/<TICKER>.md from the composite so --update can re-check it later

Modules in (italics) produce inputs, the plan, the charts or the verification; they do not get a weight in the composite.

Frameworks that do not apply to a single-stock report (run them on their own):

  • etf-analysis — if the ticker is an ETF, stop and run it instead of this report
  • portfolio-review, risk-stress-test, tax-lens — they need the user's holdings or lots
  • stock-screener, compare — they rank or weigh several tickers
  • trade-postmortem — it reviews a closed trade
  • learning-coach — run it on the finished report to have every metric explained

Research Process (5 Phases)

Modules are grouped into phases regardless of depth; each phase's sub-score is the component of the same name. Earlier phases inform later-phase assumptions.

Phase 1 — Business & Competitive Foundation → Business Quality

  • stock-eval — Company overview, financial strength, relative valuation vs. peers
  • competitor-analysis (standard+) — Moat depth, Porter's Five Forces, pricing power
  • financial-report-analyst (standard+) — Statement quality and red flags; at comprehensive depth it reads the 10k-digest first
  • industry-map (comprehensive) — Where the company sits in its value chain and who holds the pricing power
  • dividend-analysis (comprehensive) — Capital allocation: dividend and buyback sustainability, FCF deployment, M&A record, debt management

Phase 2 — Valuation → Valuation

  • stock-valuation — DCF intrinsic value with Bull/Base/Bear scenarios, plus P/E, EV/EBITDA, P/S, P/FCF peer multiples. At comprehensive depth, the bear-case downside target sets the Bear scenario

10 = deep discount to intrinsic value; 5 = fair value; 0 = extreme overvaluation.

Phase 3 — Market Signals → Market Signals

  • insider-trading — SEC Form 4 patterns, net insider sentiment
  • earnings-call-analysis (standard+) — Management tone, guidance quality, forward signals
  • institutional-ownership (standard+) — Smart-money 13F changes, concentration
  • earnings-preview (comprehensive, report within 45 days) — What the price already assumes going into the print

Phase 4 — Technical Timing → Technical Setup

  • technical-analysis — Trend structure, support/resistance, volume, momentum
  • sector-analysis (comprehensive) — Sector rotation, relative strength vs. benchmark

Phase 5 — Risk Assessment → Risk Profile

  • bear-case — Short-seller red-team: Thesis-Killers and a downside target. Its Thesis-Killers become the Bear Case section of the thesis narrative, and the Thesis Invalidation checks whether they are refuted
  • catalyst-calendar (standard+) — Dated events in the next 90 days; high-impact binary events lower the Risk Profile, and the dates populate "next catalyst dates" in the Monitoring Plan
  • short-interest, options-analysis, economics-analysis (comprehensive) — Positioning pressure, what the options market prices in, macro and rate sensitivity
  • forensic-accounting, credit-check (comprehensive) — Can the earnings be trusted, and can the balance sheet carry the plan

After the phases (comprehensive)

  • position-ladder writes the Entry Strategy; chart-master draws the charts; fact-check then result-validator verify the composite (see Data Verification above); thesis-tracker saves the thesis, its KPIs and the invalidation triggers to output/thesis/<TICKER>.md.

Composite Scoring Framework

Full Report Score = Weighted Composite

Component                Weight    Sub-Score (0-10)
─────────────────────────────────────────────────────
Business Quality          25%      [Phase 1 modules that ran]
Valuation                 25%      [Phase 2]
Market Signals            20%      [Phase 3 modules that ran]
Technical Setup           15%      [Phase 4 modules that ran]
Risk Profile              15%      [Phase 5 modules that ran]
─────────────────────────────────────────────────────
COMPOSITE SCORE          100%      X.X / 10

Composite Interpretation (the standard Score Guide — the box at the end uses the same bands):
8.0–10.0  → Strongly Bullish    (all signals aligned)
6.0–7.9   → Moderately Bullish  (most signals positive)
4.0–5.9   → Neutral             (mixed signals)
2.0–3.9   → Moderately Bearish  (mostly negative signals)
0.0–1.9   → Strongly Bearish    (strong negative signals)

The composite is the box Score, so the box Signal follows from it: ≥ 6.0 BULLISH, 4.0–5.9 NEUTRAL, < 4.0 BEARISH. Action and Conviction are a judgement on top (a 6.2 can be BULLISH / HOLD / WEAK when the valuation is stretched), but never contradict the Signal.

Sub-score derivation — a sub-score is the plain average of the box Scores of the modules in its component that ran. Every module's box Score is already oriented so that higher is better for a holder: bear-case reports min(10 − Bear Case Strength, 6.5), forensic-accounting reports 10 − Manipulation Risk Score, credit-check reports its Credit Health Score, and catalyst-calendar, short-interest, options-analysis and economics-analysis score the setup for a holder. No module is inverted a second time here.

Missing modules are never scored as 5.0. Quick depth covers each component once, so all five weights apply. If --skip (or "not applicable", e.g. earnings-preview outside the 45-day window, credit-check for a bank) removes every module of a component, drop that component and re-weight the rest proportionally — e.g. without Market Signals: Business Quality 31.25% · Valuation 31.25% · Technical Setup 18.75% · Risk Profile 18.75%. The scorecard lists every module as run / skipped by request / not applicable and states the weights actually used.


Conflict Resolution

When signals conflict across modules, apply these rules:

  • Fundamental overrides technical: Business quality and intrinsic value take precedence over short-term price action.
  • Consensus overrides outlier: When 4 of 5 phases agree on direction, document the outlier but do not let it dominate the composite score.
  • Document all conflicts explicitly: Never suppress conflicting signals. Present the full bull and bear case and explain how the weighting resolves the conflict.
  • Flag unresolvable conflicts: When signals are deeply contradictory (e.g., strong fundamentals + extreme overvaluation + heavy insider selling), flag as "Conflicted — Monitor Only" until resolution.

Investment Thesis Narrative

Every full-report output must include a narrative investment thesis structured as:

  1. Investment Thesis (2–3 paragraphs): What does the company do, why is it a compelling or poor investment now, and what is the key insight driving the thesis?
  2. Bull Case (3–5 quantifiable reasons with supporting evidence from phase outputs)
  3. Bear Case (3–5 specific risks with probability assessment and potential intrinsic value impact)
  4. Composite Score Card (visual table with all component scores, weighted scores, and interpretation)
  5. Valuation Summary (intrinsic value per share, current price, margin of safety, upside/downside % per scenario)
  6. Entry Strategy (technical setup description, ideal entry range, position sizing: Aggressive 5–7% / Moderate 2–4% / Starter 1–2%)
  7. Exit Strategy (base/bull/bear price targets, stop-loss level, time horizon)
  8. Monitoring Plan (key metrics to track quarterly, thesis-changing events, early warning signals, next catalyst dates)

HTML Report Structure

The output file is a standalone HTML document (no external dependencies except CDN Chart.js):

  • Cover page — ticker, price, date, depth flag used, overall composite score, recommendation
  • Sidebar TOC — sticky navigation linking to each section
  • Analysis sections — one per executed module plus final synthesis
  • Interactive charts — MA overlay, RSI, MACD, volume, composite score radar, phase score bar
  • Composite Score Card — weighted multi-factor scorecard with color coding
  • Signal block — standardized investment signal at report end

Design Principles

  • Clean, minimal design — white background, navy accents, Inter/system font
  • Data-dense tables with clear visual hierarchy
  • Color coding: green = bullish, amber = neutral, red = bearish
  • Print-friendly (no fixed sidebars when printing)
  • Self-contained: one .html file, no local assets

Usage

/full-report AAPL
/full-report NVDA --depth quick
/full-report MSFT --depth standard
/full-report TSLA --depth comprehensive --lang zh-TW
/full-report JPM --skip credit-check,options-analysis
/full-report AMZN --output ~/Desktop/

Arguments:

  • <TICKER> — required, any US-listed stock
  • --depth — quick | standard | comprehensive (default: comprehensive)
  • --skip — comma-separated module names to leave out at any depth; the scorecard re-weights (see Composite Scoring)
  • --lang — output language, default en, supports zh-TW
  • --output — custom save path, default output/

Output File Naming

output/<TICKER>_report_<YYYY-MM-DD>.html

Example: output/AAPL_report_2026-10-24.html


Workflow

  1. Parse --depth and --skip; determine which modules run, and mark the not-applicable ones (ETF ticker → stop and suggest etf-analysis; bank/insurer → no credit-check; next report > 45 days away → no earnings-preview)
  2. Gather all available data for <TICKER> (financials, price, filings, macro) once, as one snapshot with one as-of date; every module in the next step reads that snapshot instead of fetching its own, and a figure that is missing from it is reported as missing, not looked up again mid-run
  3. Execute modules in phase order; each phase's output informs the next
  4. Compute composite weighted score (0–10) using framework above, re-weighting for any component with no module
  5. Build investment thesis narrative with bull/bear case
  6. At comprehensive depth: Entry Strategy from position-ladder, charts from chart-master, then fact-check → result-validator on the composite
  7. Render complete HTML with all sections, charts, and scorecard
  8. Write the file to the output path, save the thesis with thesis-tracker (comprehensive), and confirm both locations


Thesis Invalidation

The composite thesis is only as good as its weakest well-weighted module. After the final synthesis, state what would reverse the composite call:

If the composite is BULLISH — thesis breaks if:

  • Any module weighted ≥ 15% flips from BULLISH to BEARISH at the next re-run (name the module and the metric that would flip it)
  • The bear-case module's Thesis-Killers are not refuted — a bull composite that cannot answer its own red-team is not bullish, it is unexamined
  • Price rises far enough that the valuation modules' fair-value range is exceeded while the fundamental modules are unchanged (multiple expansion doing all the work)

If the composite is BEARISH — thesis breaks if:

  • The specific deterioration that drove the low-scoring modules reverses (margins re-expand, guidance is raised, insider selling stops)
  • Valuation falls below the bear-case downside target with fundamentals stable (the bad news is priced)
  • Two or more independent modules (e.g. insider-trading + institutional-ownership) turn positive together

Re-run this report when:

  • Next earnings release (every module's inputs change)
  • Price moves ±15% from the level in the Data & Sources header
  • Any single module's signal flips
  • 90 days have elapsed

Output Format

End with the standard signal block after the final synthesis section:

In the output, the box below is followed immediately by this machine-readable footer — the same values as the box, filled in (no placeholders, no comments), so tools can read the result without parsing the box:

{"investskill": "1", "skill": "full-report", "ticker": "<TICKER, or null for a portfolio / market-wide view>", "as_of": "<YYYY-MM-DD from Data & Sources>", "signal": "<BULLISH | NEUTRAL | BEARISH>", "confidence": "<HIGH | MEDIUM | LOW>", "horizon": "<SHORT-TERM | MEDIUM-TERM | LONG-TERM>", "score": 0.0, "action": "<BUY | HOLD | SELL>", "conviction": "<STRONG | MODERATE | WEAK>", "retrieval": "<pasted by user | web/tool retrieval | model memory>", "sources": ["<each primary source, dated>"]}
╔══════════════════════════════════════════════╗
║              INVESTMENT SIGNAL               ║
╠══════════════════════════════════════════════╣
║ Signal:      BULLISH / NEUTRAL / BEARISH     ║
║ Confidence:  HIGH / MEDIUM / LOW             ║
║ Horizon:     SHORT / MEDIUM / LONG-TERM      ║
║ Score:       X.X / 10                        ║
╠══════════════════════════════════════════════╣
║ Action:      BUY / HOLD / SELL               ║
║ Conviction:  STRONG / MODERATE / WEAK        ║
╚══════════════════════════════════════════════╝

Score Guide: 8.0–10.0 Strongly Bullish | 6.0–7.9 Moderately Bullish | 4.0–5.9 Neutral | 2.0–3.9 Moderately Bearish | 0.0–1.9 Strongly Bearish Confidence: HIGH (strong data, clear signals) | MEDIUM (mixed signals) | LOW (limited data, conflicting signals) Horizon: SHORT-TERM (1 week–3 months) | MEDIUM-TERM (3 months–1 year) | LONG-TERM (1+ years)

Disclaimer: Educational analysis only. Not financial advice.

Fundamental Analysis

This skill has been merged into stock-eval. Use /us-stock-analysis:stock-eval for comprehensive fundamental + quality + valuation analysis.

Quick Reference — Key Ratios

RatioFormulaWhat It Measures
P/E (Price-to-Earnings)Market Price / EPSHow much investors pay per dollar of earnings; compare to peers and history
P/B (Price-to-Book)Market Price / Book Value per SharePremium over net asset value; useful for financials and asset-heavy businesses
P/S (Price-to-Sales)Market Cap / RevenueValuation relative to revenue; useful when earnings are negative or depressed
EV/EBITDAEnterprise Value / EBITDACapital-structure-neutral earnings multiple; standard M&A and cross-sector benchmark
ROE (Return on Equity)Net Income / Shareholders' EquityProfitability relative to equity book value; decompose with DuPont for quality assessment
ROIC (Return on Invested Capital)NOPAT / Invested CapitalTrue economic return on all capital deployed; compare to WACC to assess value creation
Current RatioCurrent Assets / Current LiabilitiesShort-term liquidity; >1.5x generally healthy; <1.0x signals potential liquidity stress
D/E (Debt-to-Equity)Total Debt / Shareholders' EquityFinancial leverage; higher ratio = more risk but also more return amplification

For the full analysis framework — income statement line-item breakdown, working capital cycle, DuPont decomposition, cash conversion cycle, Porter's Five Forces, Piotroski F-Score, DCF, ROIC/WACC, and risk matrix — use /us-stock-analysis:stock-eval.


Thesis Invalidation

After delivering the analysis signal, specify what would reverse it:

If signal is BULLISH — thesis breaks if:

  • revenue growth decelerates below 5% for 2 consecutive quarters AND gross margin contracts >200bps
  • [One or two more triggers drawn from this analysis's own drivers, each with a threshold]

If signal is BEARISH — thesis breaks if:

  • revenue reaccelerates >15% AND margin expansion resumes
  • [One or two more triggers drawn from this analysis's own drivers, each with a threshold]

Re-run this analysis when:

  • Next earnings release
  • Price moves ±15% from current level
  • 60 days have elapsed
  • Material news event (acquisition, leadership change, regulatory decision)

Standard Signal Output

All analysis concludes with this standardized block:

In the output, the box below is followed immediately by this machine-readable footer — the same values as the box, filled in (no placeholders, no comments), so tools can read the result without parsing the box:

{"investskill": "1", "skill": "fundamental-analysis", "ticker": "<TICKER, or null for a portfolio / market-wide view>", "as_of": "<YYYY-MM-DD from Data & Sources>", "signal": "<BULLISH | NEUTRAL | BEARISH>", "confidence": "<HIGH | MEDIUM | LOW>", "horizon": "<SHORT-TERM | MEDIUM-TERM | LONG-TERM>", "score": 0.0, "action": "<BUY | HOLD | SELL>", "conviction": "<STRONG | MODERATE | WEAK>", "retrieval": "<pasted by user | web/tool retrieval | model memory>", "sources": ["<each primary source, dated>"]}
╔══════════════════════════════════════════════╗
║              INVESTMENT SIGNAL               ║
╠══════════════════════════════════════════════╣
║ Signal:      BULLISH / NEUTRAL / BEARISH     ║
║ Confidence:  HIGH / MEDIUM / LOW             ║
║ Horizon:     SHORT / MEDIUM / LONG-TERM      ║
║ Score:       X.X / 10                        ║
╠══════════════════════════════════════════════╣
║ Action:      BUY / HOLD / SELL               ║
║ Conviction:  STRONG / MODERATE / WEAK        ║
╚══════════════════════════════════════════════╝

Score Guide: 8.0–10.0 Strongly Bullish | 6.0–7.9 Moderately Bullish | 4.0–5.9 Neutral | 2.0–3.9 Moderately Bearish | 0.0–1.9 Strongly Bearish Confidence: HIGH (strong data, clear signals) | MEDIUM (mixed signals) | LOW (limited data, conflicting signals) Horizon: SHORT-TERM (1 week–3 months) | MEDIUM-TERM (3 months–1 year) | LONG-TERM (1+ years)

Disclaimer: Educational analysis only. Not financial advice.

Industry Map — Supply & Value Chain

⚠️ Data Verification — Do This Before Any Analysis

Before running any analysis, always retrieve the latest market data for the ticker:

  1. Fetch current price — use web search or ask the user for the live price, 52-week range, and market cap. Never assume a price from training data.
  2. Confirm key figures — the revenue and gross margin of each company placed on the map (latest fiscal year), the segment that actually sits in this chain, and the source of any market-size or share figure.
  3. State your data source — fill in the Data & Sources header (next section) so the origin, as-of date, retrieval path, and confidence of every figure are explicit at the top of the output.
  4. Flag stale data explicitly — if live data is unavailable, display this warning before proceeding:

⚠️ Live data unavailable. The following analysis uses training-data estimates which may be significantly out of date. Verify all prices and metrics before making any decisions.

Never silently substitute training-data estimates for current prices. When in doubt, ask the user to paste the latest quote.


📋 Data & Sources Header — Open Every Output With It

The first thing in the output is this provenance block, filled in — never left as placeholders. It is the standard documented on the Data & Accuracy page and the first thing result-validator looks for:

Data & Sources
  As of:      <date the figures represent, e.g. 2026-06-30>
  Source:     <primary docs — SEC EDGAR 10-K/10-Q, company IR, FRED, exchange data …>
  Retrieval:  <pasted by user | web/tool retrieval | model memory>
  Confidence: <HIGH | MEDIUM | LOW>
  • Retrieval: model memory must be paired with Confidence: LOW — memory is a placeholder until confirmed against a primary source.
  • Mixed sources: list each with its own as-of date rather than blending them.
  • Sources disagree on the same figure (a closing price, a deal term): prefer the primary source (SEC filing, company release, exchange data) over an aggregator, then the most recent dated one. Show both values with their sources and dates, say which you used and why, and never average them.
  • Data the user pasted is reported as pasted by user; do not upgrade its confidence beyond what the user's own source supports.

🌐 Output Language — --lang zh-TW

English is the default. With --lang zh-TW — or when the user writes in Traditional Chinese and asks for a Chinese answer — write the whole analysis in Traditional Chinese (繁體中文, Taiwan usage), keeping it machine-checkable:

  • Translate every heading, narrative paragraph, table header and bullet. Give each metric's English name in brackets the first time it appears — 自由現金流(Free Cash Flow, FCF) — so the reader can match it to the glossary and to English analyses.
  • Keep in English tickers, figures and units, SEC form and item names (10-K Item 7, Form 4), the four Data & Sources field labels (As of, Source, Retrieval, Confidence — the Retrieval value may be written in Chinese, e.g. 使用者提供), and the Investment Signal box and JSON footer exactly as specified — field names and the BULLISH / NEUTRAL / BEARISH, HIGH / MEDIUM / LOW, BUY / HOLD / SELL, STRONG / MODERATE / WEAK values. Tools parse them.
  • Title the invalidation section 論點失效條件(Thesis Invalidation), and close with the disclaimer in Chinese: 免責聲明: 僅供教育用途,非投資建議。
  • Use Taiwan financial terms: 殖利率、本益比、營業利益率、自由現金流、聯準會、選擇權、買權/賣權、除息日、預扣稅.

Draw an industry's supply and value chain as a directed graph — from raw inputs upstream all the way down to the end user — so you can take a bird's-eye view of a business, see where a company sits in the flow, understand where value and pricing power concentrate today, and reason about where they will migrate next.

Overview

Most analysis looks at one company in isolation. This skill zooms out and answers a different question: "How does this whole industry actually work, layer by layer, and who captures the profit at each stage?"

A value chain is a directed graph: nodes are the layers of production (raw materials → components → integrators → platforms → distribution → end users), and edges point in the direction goods and services flow — from supplier to customer. Once the chain is drawn, three things become visible that a single-company view hides:

  1. Position — is the target company upstream (inputs), midstream (integration/manufacturing), or downstream (distribution / end demand)? Position determines cyclicality, margin profile, and who has pricing power over whom.
  2. Chokepoints — the layer(s) with the fewest credible suppliers capture disproportionate value. Following the chain reveals bottleneck monopolies (e.g. EUV lithography, leading-edge foundry) that are the real toll-collectors of a theme.
  3. Value migration — the profit pool is not fixed. It shifts down (or up) the chain over time as scarcity moves. Mapping the chain lets you form a thesis about where the money goes next — the "picks-and-shovels" and second-order plays.

This complements — and does not duplicate — the existing skills. competitor-analysis studies one company's moat horizontally against its direct peers; sector-analysis ranks the 11 GICS sectors for rotation. This skill maps a theme or product vertically, cutting across sectors, to show the whole flow. Output feeds naturally into competitor-analysis (pick a node, study its moat), stock-screener (rank the tickers at one layer), and chart-master / report-generator (render the graph).


1. The Value Chain Model

Every physical or digital product can be decomposed into ordered layers. Use this generic template and adapt the layer names to the specific industry:

UPSTREAM ───────────────► MIDSTREAM ───────────────► DOWNSTREAM
(scarce inputs,           (integration, manufacture,   (distribution, demand,
 tools, IP)                assembly, platforms)          the end customer)

Raw materials / inputs
   → Enabling tools & equipment
      → Components / sub-systems
         → Integrators / OEMs
            → Platforms / aggregators
               → Channel / distribution
                  → End users / demand

Node attributes — for each layer, capture:

  • Layer name and what it does in one line
  • Representative public tickers (2–5) and any dominant private players
  • Position tag: Upstream / Midstream / Downstream
  • Concentration: how many credible suppliers exist (monopoly / oligopoly / fragmented)
  • Value capture today: does this layer earn high or thin margins right now?

Edge attributes — for each arrow (supplier → customer):

  • Direction of flow (always supplier → the layer that buys from it)
  • Dependency strength (sole-source / multi-source / commodity)
  • Whether the buyer can integrate backward, or the supplier forward

2. Step 1 — Define the Scope

Clarify what is being mapped. The input is usually one of three things:

Input typeExampleWhat to map
A theme / product"AI compute", "electric vehicles", "GLP-1 drugs"The full chain end-to-end
A single tickerNVDAThe chain around it, then locate it
A layer"memory", "foundry"That layer + its immediate up/downstream neighbors

Confirm the boundaries: where does the chain start (how far upstream — mined ore? refined wafer?) and where does it end (the paying end user)? State the scope explicitly at the top of the output so the graph is bounded and legible.


3. Step 2 — Build the Chain Map (the graph)

Produce the directed graph. Default to a Mermaid flowchart (renders in Claude, Cursor, Gemini, GitHub, and the site); fall back to ASCII when Mermaid is unavailable. Hand off to chart-master for a richer HTML render or for a report-generator export.

Mermaid flowchart (primary output):

flowchart LR
    MAT["Materials & Substrates<br/>(silicon, rare earths)"] --> EQ["Fab Equipment<br/>ASML · LRCX · AMAT"]
    EQ --> FAB["Foundry / Fab<br/>TSMC · INTC · Samsung"]
    FAB --> MEM["Memory<br/>MU · Hynix · Samsung"]
    FAB --> GPU["GPU / Accelerators<br/>NVDA · AMD"]
    MEM --> GPU
    GPU --> NET["Networking / Interconnect<br/>AVGO · ANET"]
    NET --> CSP["Cloud / CSP<br/>AMZN · MSFT · GOOGL"]
    CSP --> LAB["Model Labs<br/>OpenAI · Anthropic"]
    LAB --> APP["AI Software / Apps"]
    APP --> USER["End Users<br/>enterprise · consumer"]

ASCII fallback:

Materials ─► Fab Equipment ─► Foundry ─┬─► Memory ──┐
             (ASML,LRCX,AMAT) (TSMC,INTC)   │            ▼
                                       └─► GPU/Accel (NVDA,AMD) ─► Networking (AVGO,ANET)
                                                                        │
   End Users ◄─ AI Software ◄─ Model Labs ◄─ Cloud/CSP (AMZN,MSFT) ◄────┘
   (enterprise,             (OpenAI,        (GOOGL)
    consumer)                Anthropic)

Chain map table — accompany the graph with a table so the data is machine-readable and feeds the later steps:

Layer                 Position    Key Tickers          Concentration     Value Capture (now)
──────────────────────────────────────────────────────────────────────────────────────────
Materials/Substrates  Upstream    SHECY, SUMCF         Fragmented        Low
Fab Equipment         Upstream    ASML, LRCX, AMAT     Oligopoly/Monopoly High  ◄ chokepoint
Foundry               Midstream   TSMC, INTC           Oligopoly         High
Memory                Midstream   MU, Hynix, Samsung   Oligopoly         Cyclical
GPU / Accelerators    Midstream   NVDA, AMD            Near-monopoly     Very High ◄ chokepoint
Networking            Midstream   AVGO, ANET           Oligopoly         High
Cloud / CSP           Downstream  AMZN, MSFT, GOOGL    Oligopoly         Medium (capex heavy)
Model Labs            Downstream  OpenAI, Anthropic    Fragmenting       Negative→? 
AI Software / Apps    Downstream  many                 Fragmented        Emerging
End Users             Demand      —                    —                 —

4. Step 3 — Position Locator

If the user supplied a ticker, pin it precisely:

  • Which node does it occupy? (a company may span several — e.g. Amazon is both CSP and end-market retailer)
  • Its upstream dependencies: who must it buy from, and how substitutable are they? (single-source dependency = risk)
  • Its downstream customers: who buys from it, how concentrated are they, and can they build it in-house or switch?
  • Direction of pricing power: does value flow toward this node (it can raise prices) or away (it is squeezed between a strong supplier and a strong buyer)?

State the one-line takeaway: "[Ticker] sits [upstream/mid/down] at the [layer] node; it depends on [supplier layer] and sells into [customer layer]; pricing power currently favors [node]."


5. Step 4 — Chokepoint & Bottleneck Analysis

This is where the alpha usually is. Walk the chain and score each layer for bottleneck power — the ability to hold up the entire chain:

Bottleneck Score (per layer, 0–10) = average of these four factors, each scored 0–10:
- Supplier scarcity      (fewer credible suppliers = higher)
- Substitutability       (no viable alternative = higher)
- Switching cost / lead  (longer to qualify a new supplier = higher)
- Demand inelasticity    (chain cannot proceed without it = higher)

Layers scoring high are toll collectors: they capture value regardless of who wins downstream. Classic examples to reason by analogy from: ASML (sole EUV supplier), TSMC (leading-edge foundry), NVDA + CUDA (accelerator + software lock-in). Explicitly flag:

  • Where the bottleneck is today
  • Whether it is durable or being competed / engineered away (second sources, in-housing, new architectures)
  • The "arms dealer" thesis: a bottleneck layer often wins no matter which downstream competitor prevails

6. Step 5 — Value-Pool & Margin Migration

The profit pool is dynamic. Map where margin sits now and build a thesis for where it goes next:

  • Current margin map: which layer earns the fat gross/operating margins, and which is a thin-margin commodity pass-through?
  • Scarcity shift: today's bottleneck gets competed away or over-built; a new one forms elsewhere. (e.g. compute scarcity today → energy/power and data scarcity next; hardware margins → software/services over time.)
  • Value migration direction: is value moving downstream (toward platforms and apps as hardware commoditizes) or upstream (toward whoever controls the newly scarce input)?
  • What to expect in the future: name the layer likely to capture incremental value over the next 1–3 years and the trigger that would confirm it (capacity coming online, a standard emerging, an input going scarce).

This is the section that turns a static picture into a forward-looking investment view.


7. Step 6 — Concentration & Supply-Chain Risk

A chain view exposes fragility that a single-stock view misses:

  • Single-source / single-region chokepoints (e.g. Taiwan foundry concentration, rare-earth processing in one country)
  • Geopolitical exposure: export controls, tariffs, sanctions that could sever an edge in the graph
  • Cascade risk: if one upstream node fails, how far downstream does the disruption propagate?
  • Inventory / bullwhip dynamics: demand signals amplifying up the chain (over-ordering, then a glut) — especially in memory and components
  • Customer concentration at each node: a layer selling >30% to one downstream buyer inherits that buyer's fate

8. Step 7 — Investment Idea Generation

Turn the map into a ranked idea list, organized by layer:

Layer            Best-positioned names   Rationale                         Idea type
────────────────────────────────────────────────────────────────────────────────────────
Bottleneck       [tickers]               Durable toll on the whole theme   Core / "arms dealer"
Direct winner    [tickers]               Obvious primary beneficiary        Consensus long
2nd-order        [tickers]               Sells picks-and-shovels to winners Under-covered
Squeezed         [tickers]               Caught between strong up/down      Avoid / short candidate
Optionality      [tickers]               Cheap exposure if value migrates   Speculative

Flag the non-obvious node — the second-order supplier the market under-covers because it isn't a pure-play on the theme. Hand the shortlist to stock-screener to rank, or competitor-analysis to check each name's moat, or bear-case to stress-test the consensus winner.


9. Worked Example — AI Compute Stack

Reading the map above:

  • Chokepoints: Fab Equipment (ASML EUV monopoly) and GPU/Accelerators (NVDA + CUDA). These are the durable toll collectors today.
  • Value pool now: concentrated at the GPU and foundry layers; CSPs are absorbing heavy capex; model labs are largely unprofitable.
  • Migration thesis: as accelerator supply catches up, incremental scarcity shifts toward power/energy and networking, and value accrues to software/inference that monetizes the installed base. Watch for the trigger: accelerator lead times normalizing.
  • Idea shape: core = the bottleneck arms dealers; 2nd-order = power, cooling, interconnect, and HBM memory suppliers the market under-weights; avoid = undifferentiated app-layer names with no data moat.

10. Input Formats

# Map a full theme end-to-end
/industry-map "AI compute"

# Map the chain around a ticker, then locate it
/industry-map NVDA

# Focus on one layer + its neighbors
/industry-map semiconductors --layer memory

# Emphasize where value migrates next
/industry-map "electric vehicles" --focus value-migration

# Emit chart-ready graph spec for a report
/industry-map "GLP-1 drugs" --visual

11. Visualization Support

When --visual is used, provide graph specs ready for chart-master / report-generator:

Chain Graph

Chart type: Directed graph (Mermaid flowchart LR primary, ASCII fallback, graphviz/HTML for rich export). Nodes = layers with representative tickers; edges = supplier→customer flow.

Value-Capture-by-Layer Bar

Chart type: Horizontal bar — estimated gross/operating margin (or a 0–10 value-capture score) per layer, to show where the profit pool sits.

Layer               Value-capture score (0–10)
Fab Equipment       [value]
Foundry             [value]
GPU / Accelerators  [value]
Cloud / CSP         [value]
Model Labs          [value]

Bottleneck Heat Row

Chart type: Scorecard / heat row — bottleneck score per layer, highlighting the chokepoints.


Output

Provide an industry-map report with:

  • Executive Summary (scope, where value sits today, migration thesis — 3 sentences)
  • The Chain Map (directed graph + chain-map table)
  • Position Locator (if a ticker was given)
  • Chokepoint & Bottleneck Analysis (scored, with the durable toll-collectors named)
  • Value-Pool & Margin Migration (now → next, with the confirming trigger)
  • Concentration & Supply-Chain Risk
  • Investment Ideas by layer (core / 2nd-order / avoid)
  • Investment Implications and how this feeds competitor-analysis / stock-screener / bear-case

Thesis Invalidation

After delivering the analysis signal, specify what would reverse it:

If signal is BULLISH — thesis breaks if:

  • the mapped chokepoint is broken (credible second source qualifies, or a customer in-sources) OR value migrates away from the node you favored
  • [One or two more triggers drawn from this analysis's own drivers, each with a threshold]

If signal is BEARISH — thesis breaks if:

  • the node re-establishes a durable bottleneck OR a new scarce input forms in its favor
  • [One or two more triggers drawn from this analysis's own drivers, each with a threshold]

Re-run this analysis when:

  • Next earnings release
  • Price moves ±15% from current level
  • 60 days have elapsed
  • Material news event (acquisition, leadership change, regulatory decision)

Standard Signal Output

All analysis concludes with this standardized block:

In the output, the box below is followed immediately by this machine-readable footer — the same values as the box, filled in (no placeholders, no comments), so tools can read the result without parsing the box:

{"investskill": "1", "skill": "industry-map", "ticker": "<TICKER, or null for a portfolio / market-wide view>", "as_of": "<YYYY-MM-DD from Data & Sources>", "signal": "<BULLISH | NEUTRAL | BEARISH>", "confidence": "<HIGH | MEDIUM | LOW>", "horizon": "<SHORT-TERM | MEDIUM-TERM | LONG-TERM>", "score": 0.0, "action": "<BUY | HOLD | SELL>", "conviction": "<STRONG | MODERATE | WEAK>", "retrieval": "<pasted by user | web/tool retrieval | model memory>", "sources": ["<each primary source, dated>"]}
╔══════════════════════════════════════════════╗
║              INVESTMENT SIGNAL               ║
╠══════════════════════════════════════════════╣
║ Signal:      BULLISH / NEUTRAL / BEARISH     ║
║ Confidence:  HIGH / MEDIUM / LOW             ║
║ Horizon:     SHORT / MEDIUM / LONG-TERM      ║
║ Score:       X.X / 10                        ║
╠══════════════════════════════════════════════╣
║ Action:      BUY / HOLD / SELL               ║
║ Conviction:  STRONG / MODERATE / WEAK        ║
╚══════════════════════════════════════════════╝

Score Guide: 8.0–10.0 Strongly Bullish | 6.0–7.9 Moderately Bullish | 4.0–5.9 Neutral | 2.0–3.9 Moderately Bearish | 0.0–1.9 Strongly Bearish Confidence: HIGH (strong data, clear signals) | MEDIUM (mixed signals) | LOW (limited data, conflicting signals) Horizon: SHORT-TERM (1 week–3 months) | MEDIUM-TERM (3 months–1 year) | LONG-TERM (1+ years)

Disclaimer: Educational analysis only. Not financial advice.

Insider Trading Analysis

⚠️ Data Verification — Do This Before Any Analysis

Before running any analysis, always retrieve the latest market data for the ticker:

  1. Fetch current price — use web search or ask the user for the live price, 52-week range, and market cap. Never assume a price from training data.
  2. Confirm key figures — the Form 4 filings themselves (EDGAR, filing and transaction dates, transaction code P/S/A/M/F, 10b5-1 checkbox, shares and price), and the insider's role and remaining holdings.
  3. State your data source — fill in the Data & Sources header (next section) so the origin, as-of date, retrieval path, and confidence of every figure are explicit at the top of the output.
  4. Flag stale data explicitly — if live data is unavailable, display this warning before proceeding:

⚠️ Live data unavailable. The following analysis uses training-data estimates which may be significantly out of date. Verify all prices and metrics before making any decisions.

Never silently substitute training-data estimates for current prices. When in doubt, ask the user to paste the latest quote.


📋 Data & Sources Header — Open Every Output With It

The first thing in the output is this provenance block, filled in — never left as placeholders. It is the standard documented on the Data & Accuracy page and the first thing result-validator looks for:

Data & Sources
  As of:      <date the figures represent, e.g. 2026-06-30>
  Source:     <primary docs — SEC EDGAR 10-K/10-Q, company IR, FRED, exchange data …>
  Retrieval:  <pasted by user | web/tool retrieval | model memory>
  Confidence: <HIGH | MEDIUM | LOW>
  • Retrieval: model memory must be paired with Confidence: LOW — memory is a placeholder until confirmed against a primary source.
  • Mixed sources: list each with its own as-of date rather than blending them.
  • Sources disagree on the same figure (a closing price, a deal term): prefer the primary source (SEC filing, company release, exchange data) over an aggregator, then the most recent dated one. Show both values with their sources and dates, say which you used and why, and never average them.
  • Data the user pasted is reported as pasted by user; do not upgrade its confidence beyond what the user's own source supports.

🌐 Output Language — --lang zh-TW

English is the default. With --lang zh-TW — or when the user writes in Traditional Chinese and asks for a Chinese answer — write the whole analysis in Traditional Chinese (繁體中文, Taiwan usage), keeping it machine-checkable:

  • Translate every heading, narrative paragraph, table header and bullet. Give each metric's English name in brackets the first time it appears — 自由現金流(Free Cash Flow, FCF) — so the reader can match it to the glossary and to English analyses.
  • Keep in English tickers, figures and units, SEC form and item names (10-K Item 7, Form 4), the four Data & Sources field labels (As of, Source, Retrieval, Confidence — the Retrieval value may be written in Chinese, e.g. 使用者提供), and the Investment Signal box and JSON footer exactly as specified — field names and the BULLISH / NEUTRAL / BEARISH, HIGH / MEDIUM / LOW, BUY / HOLD / SELL, STRONG / MODERATE / WEAK values. Tools parse them.
  • Title the invalidation section 論點失效條件(Thesis Invalidation), and close with the disclaimer in Chinese: 免責聲明: 僅供教育用途,非投資建議。
  • Use Taiwan financial terms: 殖利率、本益比、營業利益率、自由現金流、聯準會、選擇權、買權/賣權、除息日、預扣稅.

Comprehensive analysis of insider transaction activity to identify investment signals, patterns, and potential red flags from SEC Form 4 filings.

Analysis Framework

1. Transaction Summary (Last 6-12 Months)

Provide overview of insider trading activity:

Transaction Counts

  • Total transactions: [Number]
  • Buys: [Number] ([%] of total)
  • Sells: [Number] ([%] of total)
  • Exercises/Awards: [Number] (non-market transactions)

Transaction Volume

  • Total value traded: $[Amount]
  • Buy volume: $[Amount]
  • Sell volume: $[Amount]
  • Net insider sentiment: $[Net Amount] (Buy - Sell)

Active Period Analysis

  • Most active month: [Month]
  • Clustered activity periods: [Dates]
  • Quiet periods: [Dates]

2. Insider Sentiment Analysis

Calculate and interpret net insider sentiment:

Net Sentiment Calculation

Net Sentiment = (Buy $ - Sell $) / (Buy $ + Sell $)

Ranges:
+0.5 to +1.0  = Strongly Bullish
+0.2 to +0.5  = Moderately Bullish
-0.2 to +0.2  = Neutral
-0.5 to -0.2  = Moderately Bearish
-1.0 to -0.5  = Strongly Bearish

Sentiment Context

  • Current sentiment vs. historical baseline
  • Sentiment trend (improving/deteriorating)
  • Significance level (number and size of transactions)
  • Comparison to peer company insiders

Reliability Factors

  • Sample size (number of insiders trading)
  • Transaction size relative to holdings
  • Timing relative to information events
  • Consistency across multiple insiders

3. Significant Transactions

Highlight meaningful insider trades:

Criteria for Significance

  • Transaction size > $1 million
  • Position change > 10% of holdings
  • Unusual timing (near lows/highs, before announcements)
  • CEO or CFO transactions
  • Multiple insiders trading in same direction within short period

For Each Significant Transaction

DateInsiderPositionTransactionSharesValue% of HoldingsPriceNotes
[Date][Name][Title]Buy/Sell[#]$[Amount][%]$[Price][Context]

Transaction Analysis

  • Bullish Signals: Large insider buys, clustered buying, CEO purchases, buying near 52-week lows
  • Bearish Signals: Large insider sells, clustered selling, executive selling before bad news
  • Neutral Signals: Small routine sells for diversification, 10b5-1 plan executions

4. Insider Categories

Analyze transactions by insider role:

Executive Officers (CEO, CFO, COO)

  • Highest signal value due to information access
  • Large transactions particularly meaningful
  • Sales less concerning if part of regular compensation

Board of Directors

  • Independent directors' purchases carry strong signals
  • Less frequent trading but often meaningful
  • May have strategic visibility beyond operations

Major Shareholders (>10% owners)

  • Often have board seats or strategic influence
  • Large transactions can move markets
  • Sales may be for portfolio rebalancing

Other Insiders (VPs, Senior Managers)

  • Weaker signal but useful in aggregate
  • Clustered activity more meaningful than individual trades
  • May indicate operational knowledge

5. Transaction Types & SEC Forms

Interpret different transaction codes:

Form 4 Transaction Codes

  • P (Purchase): Open market buy - BULLISH SIGNAL
  • S (Sale): Open market sell - BEARISH SIGNAL (but often routine)
  • M (Exercise): Option exercise - Neutral
  • F (Tax Payment): Shares surrendered for taxes - Neutral
  • A (Award): Stock grant/award - Neutral
  • G (Gift): Gift transfer - Neutral
  • D (Disposition): Return to company - Neutral

Form 3 vs. Form 4 vs. Form 5

  • Form 3: Initial statement of beneficial ownership (new insider)
  • Form 4: Changes in ownership (filed within 2 business days)
  • Form 5: Annual statement of changes (deferred reporting)

10b5-1 Trading Plans

  • Pre-arranged trading plans to avoid insider trading allegations
  • Marked in filings if transaction was under 10b5-1 — since April 2023, Form 4 has a dedicated checkbox and the adoption date goes in a footnote
  • The SEC's 2023 amendments added a cooling-off period before the first trade (directors and officers: the later of 90 days after adoption or two business days after the next 10-Q/10-K, capped at 120 days; others: 30 days), limit single-trade plans to one per 12 months, and bar overlapping plans; issuers now disclose officer and director plan adoptions and terminations each quarter (10-Q/10-K, Item 408)
  • Less signal value than discretionary trades
  • Look for plan adoption/termination patterns — an adoption or termination shortly before material news is the informative event, and the quarterly Item 408 table is where to find it

6. Ownership Trends

Track changes in insider ownership percentage:

Current Ownership Structure

  • Total insider ownership: [%]
  • CEO ownership: [%]
  • Top 5 insiders combined: [%]

Ownership Trend (12 months)

Quarter        Total Insider %    Change from Prior
[Qn YYYY]          [%]              [±%]
[Qn-1 YYYY]        [%]              [±%]
[Qn-2 YYYY]        [%]              [±%]
[Qn-3 YYYY]        [%]              [±%]

Interpretation

  • Increasing ownership: Insiders accumulating = Bullish
  • Stable ownership: Confidence maintained = Neutral to Bullish
  • Decreasing ownership: Distribution = Bearish (but check reasons)

Skin in the Game Assessment

  • High insider ownership (>20%): Strong alignment
  • Medium ownership (10-20%): Adequate alignment
  • Low ownership (<10%): Weaker alignment
  • CEO ownership relative to compensation

7. Timing Analysis

Evaluate timing of insider transactions:

Relative to Stock Price

  • Buying near 52-week lows: Strong bullish signal
  • Buying after significant decline: Contrarian confidence
  • Selling near 52-week highs: Less bearish (profit-taking)
  • Selling before decline: Red flag if not disclosed later

Relative to Information Events

  • Before earnings release
  • Before major announcements (M&A, product launches)
  • During blackout periods (should be restricted)
  • After significant news (reaction to public information)

Seasonal Patterns

  • Post-earnings window trading (common)
  • Year-end tax-loss selling
  • Quarterly option exercise cycles
  • Annual equity grant patterns

Red Flag Timing

  • Trading during supposed blackout periods
  • Selling shortly before bad news announcement
  • Exercising and immediately selling (vs. holding)
  • Rapid succession of sells by multiple insiders

8. Pattern Recognition

Identify meaningful patterns in trading activity:

Bullish Patterns

  • Clustered buying by multiple insiders within 1-2 weeks
  • First-time purchases by long-tenured executives
  • Buying during period of stock weakness or negative sentiment
  • Insiders increasing position size significantly (>25%)
  • CEO making largest purchase of tenure

Bearish Patterns

  • Clustered selling by multiple insiders
  • CEO or CFO selling large percentage of holdings
  • First-time sells by insiders who previously only bought
  • Selling despite stock decline (urgency to exit)
  • Executive departures followed by sales

Neutral Patterns

  • Regular quarterly option exercises and sells
  • Small percentage sells for diversification (<5% of holdings)
  • Tax payment share forfeitures
  • Routine 10b5-1 plan executions
  • Gift transfers to family trusts

9. Red Flags & Warning Signs

Identify concerning patterns:

High-Severity Red Flags

  • CEO or CFO selling >50% of holdings
  • Clustered selling before earnings miss or guidance cut
  • Trading during blackout window
  • Late Form 4 filing (>2 days) without explanation
  • Pattern of selling before negative news

Medium-Severity Red Flags

  • Increasing rate of insider selling
  • Multiple executives selling simultaneously
  • Selling despite recent insider buying by others
  • Large sales shortly after option vesting
  • Board member resignations followed by sales

Context Matters

  • Insider selling is common and often benign
  • Check for 10b5-1 plans (pre-arranged sales)
  • Consider personal circumstances (divorce, estate planning)
  • Look at historical patterns for that individual
  • Assess whether selling is proportional to holdings

10. Data Sources & Research Tools

Reference the following sources for insider trading data:

Primary Source - SEC EDGAR

Aggregator Websites

  • OpenInsider.com: Free, comprehensive database with filters

    • Cluster analysis tools
    • Screeners for significant transactions
    • Historical data back 10+ years
  • SEC Form 4 Tracker: Real-time Form 4 monitoring

  • Finviz Insider: Basic insider data with visualizations

  • GuruFocus Insider: Premium insider analysis

Premium Tools

  • Bloomberg Terminal: Comprehensive insider data and analytics
  • FactSet: Insider transaction database
  • S&P Capital IQ: Detailed ownership and transaction data

Key Data Points to Extract

  • Insider name and title
  • Transaction date and filing date
  • Shares bought/sold
  • Price per share
  • Shares owned after transaction
  • Transaction code (P, S, M, etc.)
  • Form 4 footnotes (10b5-1 plan indicator)

Input Formats

Format 1: Ticker Symbol

User: /insider-trading AAPL

Claude retrieves recent Form 4 data from SEC EDGAR and analyzes

Format 2: Specific Time Period

User: /insider-trading TSLA --period 2024-Q4

Claude focuses on Q4 2024 insider transactions

Format 3: Custom Data Paste

User: /insider-trading --data

[Paste Form 4 data table or CSV from OpenInsider/SEC]

Claude analyzes provided data

Output

Provide comprehensive insider trading analysis report with:

1. Executive Summary

  • Overall Insider Sentiment: Bullish / Neutral / Bearish
  • Confidence Level: High / Medium / Low
  • Key Takeaways: 3-5 most important findings
  • Investment Signal: Buy / Hold / Sell indication
  • Time Period Analyzed: [Date range]

2. Transaction Summary Table

Period: Last 6 Months
Total Transactions: [N]
Total Buy Transactions: [N] ($[Amount])
Total Sell Transactions: [N] ($[Amount])
Net Insider Activity: $[Amount]
Net Sentiment Score: [+0.XX or -0.XX]

3. Significant Transactions (Top 10)

Detailed table with context for each major transaction

4. Insider Category Breakdown

Analysis by executive officers, board members, and major shareholders

5. Ownership Trend Analysis

Current ownership % and historical trend (4 quarters)

6. Timing & Pattern Analysis

  • Stock price context during trading periods
  • Identified bullish or bearish patterns
  • Seasonal or cyclical factors

7. Red Flags & Concerns

  • List any identified warning signs
  • Severity assessment
  • Recommended monitoring

8. Comparison to Peers

Brief comparison of insider sentiment vs. industry peers (if available)

9. Historical Context

  • How current activity compares to historical baseline
  • Notable changes in behavior
  • Track record of insider signals

10. Investment Implications

  • Overall assessment and confidence level
  • Key bullish factors from insider activity
  • Key bearish factors or concerns
  • Recommended action
  • Monitoring plan for future activity

11. Data Sources Used

  • SEC EDGAR filings referenced
  • Date range of analysis
  • Number of Form 4 filings reviewed

Interpretation Guidelines

When Insider Buying is STRONGLY Bullish

  1. Multiple insiders buying in 1-2 week period
  2. CEO making unusually large purchase
  3. Buying after significant stock decline (>20%)
  4. Insiders buying with personal cash (not option exercises)
  5. First-time buyers or return to buying after long gap

When Insider Buying is MODERATELY Bullish

  1. Single insider making meaningful purchase
  2. Regular accumulation pattern
  3. Buying at all-time highs (conviction in further growth)
  4. Board member purchases

When Insider Selling is LESS Concerning

  1. 10b5-1 pre-arranged plan sales
  2. Small percentage of holdings (<5%)
  3. Regular quarterly pattern
  4. Diversification after large stock appreciation
  5. Tax payments or estate planning

When Insider Selling is CONCERNING

  1. Clustered selling by multiple executives
  2. CEO/CFO selling >25% of holdings
  3. Selling despite stock decline
  4. No 10b5-1 plan indicated
  5. Selling shortly before negative news

Signal Reliability Factors

  • Sample size: More insiders trading = stronger signal
  • Magnitude: Larger $ amounts = stronger signal
  • Timing: Unusual timing = stronger signal
  • Context: Contrarian trades (buying weakness) = stronger signal
  • Track record: Historically accurate insiders = stronger signal

Usage Examples

Example 1: Basic Analysis

User: /insider-trading NVDA

Claude: Pulls the last 6 months of NVDA Form 4 filings and reports net
open-market buys vs. sells, separates 10b5-1 plan sales from discretionary
trades, flags clustered buying, and closes with the signal block.

Example 2: Before Earnings

User: /insider-trading AAPL --period "before-earnings"

Claude: Examines trading in the 30–60 days before the most recent earnings
release and states whether anything beyond scheduled 10b5-1 sales occurred.

Example 3: Red Flag Detection

User: /insider-trading ABC --focus red-flags

Claude: Checks each red-flag pattern above (large % of holdings sold,
clustered director exits, selling into a decline) and reports which fired,
with the filing dates that support each.

Integration Notes

  • Combine with /stock-eval for complete due diligence
  • Use /institutional-ownership for full insider + institutional picture
  • Feed analysis to /report-generator for visual charts of insider activity
  • Best used monthly or after significant insider filing activity
  • Particularly valuable when analyzing small-mid cap stocks (insiders have more edge)

Thesis Invalidation

After delivering the analysis signal, specify what would reverse it:

If signal is BULLISH — thesis breaks if:

  • 3+ insiders file Form 4 sells within a 30-day window OR CEO sells >5% of holdings
  • [One or two more triggers drawn from this analysis's own drivers, each with a threshold]

If signal is BEARISH — thesis breaks if:

  • CEO/CFO initiate new purchases >$1M within 30 days
  • [One or two more triggers drawn from this analysis's own drivers, each with a threshold]

Re-run this analysis when:

  • Next earnings release
  • Price moves ±15% from current level
  • 60 days have elapsed
  • Material news event (acquisition, leadership change, regulatory decision)

Standard Signal Output

All analysis concludes with this standardized block:

In the output, the box below is followed immediately by this machine-readable footer — the same values as the box, filled in (no placeholders, no comments), so tools can read the result without parsing the box:

{"investskill": "1", "skill": "insider-trading", "ticker": "<TICKER, or null for a portfolio / market-wide view>", "as_of": "<YYYY-MM-DD from Data & Sources>", "signal": "<BULLISH | NEUTRAL | BEARISH>", "confidence": "<HIGH | MEDIUM | LOW>", "horizon": "<SHORT-TERM | MEDIUM-TERM | LONG-TERM>", "score": 0.0, "action": "<BUY | HOLD | SELL>", "conviction": "<STRONG | MODERATE | WEAK>", "retrieval": "<pasted by user | web/tool retrieval | model memory>", "sources": ["<each primary source, dated>"]}
╔══════════════════════════════════════════════╗
║              INVESTMENT SIGNAL               ║
╠══════════════════════════════════════════════╣
║ Signal:      BULLISH / NEUTRAL / BEARISH     ║
║ Confidence:  HIGH / MEDIUM / LOW             ║
║ Horizon:     SHORT / MEDIUM / LONG-TERM      ║
║ Score:       X.X / 10                        ║
╠══════════════════════════════════════════════╣
║ Action:      BUY / HOLD / SELL               ║
║ Conviction:  STRONG / MODERATE / WEAK        ║
╚══════════════════════════════════════════════╝

Score Guide: 8.0–10.0 Strongly Bullish | 6.0–7.9 Moderately Bullish | 4.0–5.9 Neutral | 2.0–3.9 Moderately Bearish | 0.0–1.9 Strongly Bearish Confidence: HIGH (strong data, clear signals) | MEDIUM (mixed signals) | LOW (limited data, conflicting signals) Horizon: SHORT-TERM (1 week–3 months) | MEDIUM-TERM (3 months–1 year) | LONG-TERM (1+ years)

Disclaimer: Educational analysis only. Not financial advice.

Institutional Ownership Analysis

⚠️ Data Verification — Do This Before Any Analysis

Before running any analysis, always retrieve the latest market data for the ticker:

  1. Fetch current price — use web search or ask the user for the live price, 52-week range, and market cap. Never assume a price from training data.
  2. Confirm key figures — the 13F quarter-end date (filings arrive up to 45 days later, so holdings can be ~4 months old), shares and value per manager, quarter-over-quarter change, and total institutional ownership %.
  3. State your data source — fill in the Data & Sources header (next section) so the origin, as-of date, retrieval path, and confidence of every figure are explicit at the top of the output.
  4. Flag stale data explicitly — if live data is unavailable, display this warning before proceeding:

⚠️ Live data unavailable. The following analysis uses training-data estimates which may be significantly out of date. Verify all prices and metrics before making any decisions.

Never silently substitute training-data estimates for current prices. When in doubt, ask the user to paste the latest quote.


📋 Data & Sources Header — Open Every Output With It

The first thing in the output is this provenance block, filled in — never left as placeholders. It is the standard documented on the Data & Accuracy page and the first thing result-validator looks for:

Data & Sources
  As of:      <date the figures represent, e.g. 2026-06-30>
  Source:     <primary docs — SEC EDGAR 10-K/10-Q, company IR, FRED, exchange data …>
  Retrieval:  <pasted by user | web/tool retrieval | model memory>
  Confidence: <HIGH | MEDIUM | LOW>
  • Retrieval: model memory must be paired with Confidence: LOW — memory is a placeholder until confirmed against a primary source.
  • Mixed sources: list each with its own as-of date rather than blending them.
  • Sources disagree on the same figure (a closing price, a deal term): prefer the primary source (SEC filing, company release, exchange data) over an aggregator, then the most recent dated one. Show both values with their sources and dates, say which you used and why, and never average them.
  • Data the user pasted is reported as pasted by user; do not upgrade its confidence beyond what the user's own source supports.

🌐 Output Language — --lang zh-TW

English is the default. With --lang zh-TW — or when the user writes in Traditional Chinese and asks for a Chinese answer — write the whole analysis in Traditional Chinese (繁體中文, Taiwan usage), keeping it machine-checkable:

  • Translate every heading, narrative paragraph, table header and bullet. Give each metric's English name in brackets the first time it appears — 自由現金流(Free Cash Flow, FCF) — so the reader can match it to the glossary and to English analyses.
  • Keep in English tickers, figures and units, SEC form and item names (10-K Item 7, Form 4), the four Data & Sources field labels (As of, Source, Retrieval, Confidence — the Retrieval value may be written in Chinese, e.g. 使用者提供), and the Investment Signal box and JSON footer exactly as specified — field names and the BULLISH / NEUTRAL / BEARISH, HIGH / MEDIUM / LOW, BUY / HOLD / SELL, STRONG / MODERATE / WEAK values. Tools parse them.
  • Title the invalidation section 論點失效條件(Thesis Invalidation), and close with the disclaimer in Chinese: 免責聲明: 僅供教育用途,非投資建議。
  • Use Taiwan financial terms: 殖利率、本益比、營業利益率、自由現金流、聯準會、選擇權、買權/賣權、除息日、預扣稅.

Comprehensive analysis of institutional investor holdings to identify "smart money" trends, ownership concentration, and investment signals from SEC 13F filings.

Analysis Framework

1. Institutional Ownership Overview

Provide current snapshot of institutional holdings:

Aggregate Statistics

  • Total institutional ownership: [%] of shares outstanding
  • Number of institutional holders: [N]
  • Total shares held by institutions: [Number] shares
  • Total value held: $[Amount]
  • Float ownership by institutions: [%] (excl. insider holdings)

Ownership Trend (4 Quarters)

Quarter        Institutional %    # of Holders    Change from Prior
[Qn YYYY]          [%]                [N]              [±%]
[Qn-1 YYYY]        [%]                [N]              [±%]
[Qn-2 YYYY]        [%]                [N]              [±%]
[Qn-3 YYYY]        [%]                [N]              [±%]

Trend Interpretation

  • Increasing ownership: Growing institutional interest = Bullish
  • Stable ownership: Maintained confidence = Neutral to Bullish
  • Decreasing ownership: Distribution or loss of confidence = Bearish

2. Top Institutional Holders

Identify and analyze largest institutional positions:

Top 10 Holders Table

RankInstitutionTypeShares HeldValue ($M)% of Portfolio% of CompanyChange (QoQ)
1Vanguard GroupIndex45.2M$6,7800.8%8.5%+2.1%
2BlackRockIndex38.1M$5,7150.6%7.2%+1.5%
3State StreetIndex22.3M$3,3451.2%4.2%-0.3%
4FidelityActive18.7M$2,8052.1%3.5%+15.2%
5.....................

Holder Categories

  • Index Funds (Vanguard, BlackRock, State Street): Follow index weights, less signal
  • Active Managers (Fidelity, T. Rowe Price, Wellington): Stock selection, more signal
  • Hedge Funds (Bridgewater, Citadel, Renaissance): Tactical, high signal value
  • Pension Funds (CalPERS, CalSTRS): Long-term, moderate signal
  • Sovereign Wealth Funds: Very long-term, strategic positions

3. Quarter-over-Quarter Position Changes

Analyze recent changes in institutional holdings:

New Positions (Initiated)

  • Institutions that added stock for first time in recent quarter
  • Particularly meaningful if from notable investors
  • Indicates new conviction or thesis development
InstitutionSharesValue ($M)% of Their PortfolioSignificance
ARK Invest2.5M$3753.2%High (active, thematic)
...............

Increased Positions (>25% increase)

  • Institutions that significantly added to existing position
  • Shows growing conviction
  • More meaningful if they're "adding to winners"
InstitutionPrevious SharesNew SharesChangeNew ValueNotes
ValueAct Capital5.0M8.2M+64%$1,230MActivist, 4th increase
..................

Decreased Positions (>25% decrease)

  • Institutions that significantly reduced holdings
  • May indicate profit-taking, thesis change, or risk management
  • Consider whether selling from highs or lows
InstitutionPrevious SharesNew SharesChangeCurrent ValueReason
Tiger Global12.0M4.5M-62%$675MReducing tech exposure
..................

Eliminated Positions (Sold Completely)

  • Complete exits by institutional holders
  • Most bearish signal, especially if from concentrated holders
  • Check if stock was top holding previously
InstitutionPrevious SharesValue Sold ($M)Exit PriceContext
Scion Asset Mgmt3.2M$480$150Burry exited
...............

4. Smart Money Analysis

Track notable investors and their activities:

Super Investors / Gurus (Defined as legendary investors or top-performing fund managers)

Bullish Signals - Notable Buyers

InvestorInstitutionActionSharesValue ($M)% of Their PortfolioSignal Strength
Warren BuffettBerkshire HathawayNew Position50.0M$7,5004.5%VERY HIGH
Bill AckmanPershing SquareIncreased 80%8.5M$1,2758.2%HIGH
.....................

Bearish Signals - Notable Sellers

InvestorInstitutionActionShares SoldValue ($M)Reason (if stated)Signal Strength
Michael BurryScion Asset MgmtEliminated3.2M$480Not disclosedHIGH
.....................

Notable Investor Categories

  1. Value Investors: Buffett, Klarman, Pabrai, Greenblatt
  2. Growth Investors: T. Rowe Price, Baron, Baillie Gifford
  3. Activist Investors: Icahn, Ackman, Loeb, ValueAct
  4. Quant/Macro: Bridgewater, Renaissance, AQR, Two Sigma
  5. Tech-Focused: ARK Invest, Tiger Global, Coatue

Why Smart Money Matters

  • Superior research resources and information access
  • Proven track records of alpha generation
  • Often catalyst for stock re-rating
  • Activist investors can drive corporate changes
  • Attracts other institutional followers

5. Ownership Concentration Analysis

Assess how concentrated institutional ownership is:

Concentration Metrics

  • Top 3 holders: [%] of shares outstanding
  • Top 10 holders: [%] of shares outstanding
  • Top 25 holders: [%] of shares outstanding
  • HHI (Herfindahl-Hirschman Index): [Score] (concentration measure)

Interpretation

  • High concentration (Top 10 > 50%):

    • Pro: Strong conviction from major holders
    • Con: Volatile if large holder exits
    • Common in: Small-caps, founder-led companies
  • Medium concentration (Top 10 = 30-50%):

    • Balanced ownership structure
    • Most large-cap companies
  • Low concentration (Top 10 < 30%):

    • Widely distributed ownership
    • More stable, less influenced by single holder
    • Common in: Mega-caps, index heavyweights

Turnover Analysis

  • Holder retention rate: [%] of holders from 1 year ago still holding
  • New entrants (last 12 months): [N] institutions
  • Complete exits (last 12 months): [N] institutions
  • Net holder change: [+/-N]

Stability Assessment

  • High retention + new entrants = Strengthening conviction (Bullish)
  • High retention + no new entrants = Stable (Neutral)
  • Low retention + exits = Weakening conviction (Bearish)

6. Activist & Strategic Holders

Identify and analyze activist investors or strategic stakeholders:

Activist Investors (>5% ownership or known activist activity)

For each activist position:

  • Investor Name & Fund
  • Ownership %: Current stake
  • Entry Date & Price: When position was initiated
  • Current Value: Mark-to-market
  • Activist Thesis: What changes they're pushing for
  • Progress: Status of their initiatives
  • Track Record: Success rate in similar situations

Types of Activist Campaigns

  1. Operational Improvements: Cost cuts, margin expansion, divestitures
  2. Capital Allocation: Increased buybacks, dividends, M&A
  3. Strategic Alternatives: Sale of company, merger, spin-offs
  4. Board Changes: New directors, CEO replacement
  5. Governance: ESG initiatives, compensation reforms

Impact Assessment

  • Historical success rate of this activist at peer companies
  • Management's receptiveness to proposals
  • Other shareholders' support
  • Potential value unlock if successful
  • Timeline to value realization

7. Portfolio Weight Analysis

Analyze how important the position is to each holder:

High-Conviction Holders (Position is >5% of their portfolio)

InstitutionSharesValue ($M)% of Their PortfolioRank in PortfolioRecent Action
Akre Capital4.2M$6308.5%#2 holdingIncreased
..................

Interpretation

  • High portfolio weight: Strong conviction, more research depth
  • Top 10 holding: Likely actively managed and monitored
  • Increases to high-conviction holding: Very bullish signal
  • Decreases from high-conviction: Concerning if thesis change

Position Sizing Trends

  • How many institutions have it as top 10 holding: [N]
  • How many increased portfolio weight: [N] (bullish)
  • How many decreased portfolio weight: [N] (bearish)
  • Average portfolio weight: [%]

8. Stock Price Performance Correlation

Analyze relationship between ownership changes and stock performance:

Performance During Institutional Activity

Period: [Qn YYYY]
Institutional Change: [±%] ownership (net buying / selling)
Stock Price Change: [±%]
S&P 500 Change: [±%]
Outperformance: [±%]

Historical Pattern Analysis

  • Correlation between institutional buying and subsequent returns
  • Lead/lag relationship (do institutions front-run or follow?)
  • Performance after smart money buys vs. sells

Predictive Value

  • High predictive value scenarios:

    • Multiple smart money managers buying
    • Activist initiating position
    • Institutions buying during weakness
    • First-time institutional coverage
  • Low predictive value scenarios:

    • Index rebalancing flows
    • ETF creation/redemption
    • Tax-loss selling end-of-year
    • Merger arbitrage activity

9. Data Sources & Filing Requirements

Reference the following sources for institutional ownership data:

Primary Source - SEC 13F Filings

  • What: Quarterly holdings reports required of institutional investment managers exercising investment discretion over ≥ $100 million in Section 13(f) securities (the test is 13(f) securities under management, not the firm's total AUM)
  • Who: Institutional investment managers
  • When: Filed within 45 days of quarter-end
  • Where: https://www.sec.gov/cgi-bin/browse-edgar?action=getcompany&type=13F

13F Filing Calendar

  • Q1 (Mar 31) → Filed by May 15
  • Q2 (Jun 30) → Filed by Aug 14
  • Q3 (Sep 30) → Filed by Nov 14
  • Q4 (Dec 31) → Filed by Feb 14

Important Limitations

  • 45-day lag (data is stale)
  • Only long positions reported (shorts not disclosed)
  • Options and ADRs may not be reported
  • Some holdings can be confidentially withheld
  • $200K or 10,000 share threshold per position

Schedule 13D/G Filings (>5% ownership)

  • 13D: Activist investors with intent to influence

    • Initial filing within 5 business days of crossing 5% (SEC rule effective 2024-02-05; previously 10 calendar days)
    • Amendments for material changes (≥ 1%) within 2 business days
    • Includes plans and intentions
  • 13G: Passive investors, no control intent

    • Deadlines depend on filer type (revised schedule in force since 2024-09-30): qualified institutional investors — 45 days after the quarter-end in which they cross 5%, or 5 business days after month-end once above 10%; passive investors — 5 business days after crossing 5%; exempt investors — 45 days after quarter-end
    • Verify the current deadline on sec.gov before dating a filing as late
    • Simplified reporting for passive positions

Aggregator Services

  • WhaleWisdom.com: Free 13F database with analytics
  • Dataroma: Tracks super investor portfolios
  • GuruFocus: Premium institutional tracking
  • Whalewisdom: Advanced screening and alerts
  • Bloomberg/FactSet: Comprehensive institutional data (paid)

10. Red Flags & Warning Signs

Identify concerning patterns in institutional ownership:

High-Severity Red Flags

  • Multiple smart money managers exiting simultaneously
  • Decreasing ownership despite rising stock price (selling into strength)
  • Activist investor exiting after failed campaign
  • Large holders selling >50% of position
  • Net institutional selling for 3+ consecutive quarters

Medium-Severity Red Flags

  • Increasing number of holder exits (even if small positions)
  • Declining portfolio weights across multiple holders
  • No new institutional buyers despite attractive valuation
  • High turnover among holders (low retention)

Positive Signals

  • Smart money buying during market selloff
  • New positions from value-oriented long-term holders
  • Activist entering with specific value-creation plan
  • Increasing high-conviction holdings (>5% of portfolio)
  • Multiple quarters of consistent net institutional buying

Input Formats

Format 1: Ticker Symbol

User: /institutional-ownership AAPL

Claude retrieves latest 13F data and analyzes recent changes

Format 2: Specific Holder Tracking

User: /institutional-ownership TSLA --track "Berkshire,ARK,Scion"

Claude focuses on specific institutional holders' activity

Format 3: Custom Data Analysis

User: /institutional-ownership --data

[Paste 13F data table from WhaleWisdom or SEC]

Claude analyzes provided institutional holdings data

Output

Provide comprehensive institutional ownership analysis report with:

1. Executive Summary

  • Overall Institutional Sentiment: Bullish / Neutral / Bearish
  • Confidence Level: High / Medium / Low
  • Key Findings: 3-5 most important insights
  • Investment Signal: Buy / Hold / Sell indication
  • Smart Money Direction: Net buying / selling

2. Ownership Overview

  • Current institutional ownership %
  • Trend over last 4 quarters
  • Number of holders and change
  • Total value held

3. Top 10 Holders

Detailed table with position sizes, changes, and portfolio weights

4. Recent Activity Summary

  • New positions count and notable investors
  • Increased positions (>25% increases)
  • Decreased positions (>25% decreases)
  • Eliminated positions

5. Smart Money Tracker

Notable investors' activity with signal strength assessment

6. Ownership Concentration

  • Top 3/10/25 holder concentration
  • Turnover and stability metrics
  • Assessment of ownership structure

7. Activist & Strategic Holdings

Any activist investors or strategic stakeholders (>5% ownership)

8. High-Conviction Holders

Institutions with >5% portfolio weight in this stock

9. Performance Correlation

Stock performance during periods of institutional buying/selling

10. Red Flags & Positive Signals

  • Warning signs identified
  • Bullish indicators present
  • Severity assessment

11. Investment Implications

  • Overall assessment and confidence
  • Key bullish factors from institutional data
  • Key bearish factors or concerns
  • Recommended action
  • Key levels/catalysts to monitor

12. Data Sources & Timing

  • 13F filing quarter analyzed
  • Number of 13F filings reviewed
  • Any data limitations or gaps

Interpretation Guidelines

When Institutional Buying is STRONGLY Bullish

  1. Smart money managers initiating or significantly increasing (>50%)
  2. Multiple value investors buying same stock
  3. Activist investor initiating with specific value-creation plan
  4. Net institutional buying for 3+ consecutive quarters
  5. Buying during market weakness (contrarian signal)

When Institutional Buying is MODERATELY Bullish

  1. Increasing ownership trend (1-2 quarters)
  2. Growing number of institutional holders
  3. New positions from credible active managers
  4. Increasing portfolio weights across multiple holders

When Institutional Selling is LESS Concerning

  1. Index fund rebalancing (mechanical selling)
  2. Profit-taking after significant appreciation
  3. Small position eliminations (<1% of portfolio)
  4. Single quarter of selling after long accumulation
  5. Sector rotation or style drift

When Institutional Selling is CONCERNING

  1. Smart money managers exiting completely
  2. Activist investors giving up and selling
  3. Selling during stock decline (lack of confidence)
  4. Multiple quarters of net institutional selling
  5. Decreasing number of holders

Signal Reliability Factors

  • Quality of holders: Smart money > Index funds
  • Conviction level: High portfolio weights = stronger signal
  • Consistency: Multiple quarters of buying/selling = stronger
  • Context: Buying weakness or selling strength = stronger signal
  • Magnitude: Large position changes = stronger signal

Usage Examples

Example 1: Basic Analysis

User: /institutional-ownership NVDA

Claude: Reads the latest quarter's 13F filings and reports the change in
institutional ownership, new and exited holders, and the largest adds and
trims by named manager, then closes with the signal block.

Example 2: Smart Money Tracking

User: /institutional-ownership META --smart-money

Claude: Restricts the holder analysis to the smart-money list above and
reports each manager's position change for the quarter.

Example 3: Comparative Analysis

User: /institutional-ownership AAPL MSFT GOOGL --compare

Claude: Runs the same ownership-trend and smart-money checks on each ticker
and ranks them by net institutional accumulation.

Integration Notes

  • Combine with /insider-trading for complete ownership analysis
  • Use with /stock-eval for comprehensive due diligence
  • Feed to /report-generator for visual ownership trend charts
  • Best used quarterly after 13F filing deadlines (May, Aug, Nov, Feb)
  • Particularly valuable for small-mid cap stocks where institutional buying can be catalyst
  • Less meaningful for mega-caps where ownership changes are primarily index-driven

Thesis Invalidation

After delivering the analysis signal, specify what would reverse it:

If signal is BULLISH — thesis breaks if:

  • top 3 holders reduce positions by >20% in a single quarter
  • [One or two more triggers drawn from this analysis's own drivers, each with a threshold]

If signal is BEARISH — thesis breaks if:

  • 2+ top-tier institutions (BlackRock, Vanguard, Fidelity) initiate new positions
  • [One or two more triggers drawn from this analysis's own drivers, each with a threshold]

Re-run this analysis when:

  • Next earnings release
  • Price moves ±15% from current level
  • 60 days have elapsed
  • Material news event (acquisition, leadership change, regulatory decision)

Standard Signal Output

All analysis concludes with this standardized block:

In the output, the box below is followed immediately by this machine-readable footer — the same values as the box, filled in (no placeholders, no comments), so tools can read the result without parsing the box:

{"investskill": "1", "skill": "institutional-ownership", "ticker": "<TICKER, or null for a portfolio / market-wide view>", "as_of": "<YYYY-MM-DD from Data & Sources>", "signal": "<BULLISH | NEUTRAL | BEARISH>", "confidence": "<HIGH | MEDIUM | LOW>", "horizon": "<SHORT-TERM | MEDIUM-TERM | LONG-TERM>", "score": 0.0, "action": "<BUY | HOLD | SELL>", "conviction": "<STRONG | MODERATE | WEAK>", "retrieval": "<pasted by user | web/tool retrieval | model memory>", "sources": ["<each primary source, dated>"]}
╔══════════════════════════════════════════════╗
║              INVESTMENT SIGNAL               ║
╠══════════════════════════════════════════════╣
║ Signal:      BULLISH / NEUTRAL / BEARISH     ║
║ Confidence:  HIGH / MEDIUM / LOW             ║
║ Horizon:     SHORT / MEDIUM / LONG-TERM      ║
║ Score:       X.X / 10                        ║
╠══════════════════════════════════════════════╣
║ Action:      BUY / HOLD / SELL               ║
║ Conviction:  STRONG / MODERATE / WEAK        ║
╚══════════════════════════════════════════════╝

Score Guide: 8.0–10.0 Strongly Bullish | 6.0–7.9 Moderately Bullish | 4.0–5.9 Neutral | 2.0–3.9 Moderately Bearish | 0.0–1.9 Strongly Bearish Confidence: HIGH (strong data, clear signals) | MEDIUM (mixed signals) | LOW (limited data, conflicting signals) Horizon: SHORT-TERM (1 week–3 months) | MEDIUM-TERM (3 months–1 year) | LONG-TERM (1+ years)

Disclaimer: Educational analysis only. Not financial advice.

Learning Coach — Understand the Analysis You Just Got

⚠️ Data Verification — Do This Before Any Analysis

Before explaining anything, check the material you were handed:

  1. Do not re-fetch data — explain the figures as they appear in the pasted analysis, and flag any that look stale (as-of date older than 90 days), unsourced (no Data & Sources header), or internally inconsistent (a score that does not match its own signal band). Teaching from bad numbers teaches bad habits.
  2. Confirm key figures — where the pasted analysis states a metric without its inputs (a P/E with no price or EPS), say so; do not invent the missing inputs to make the explanation tidier.
  3. State your data source — fill in the Data & Sources header (next section) so the origin, as-of date, retrieval path, and confidence of every figure are explicit at the top of the output.
  4. Flag stale data explicitly — if the pasted analysis carries a live-data warning or says Retrieval: model memory, display this before proceeding:

⚠️ Live data unavailable. The following analysis uses training-data estimates which may be significantly out of date. Verify all prices and metrics before making any decisions.

If the pasted analysis simply has no as-of date, do not assume it came from memory: write As of: unknown in the reproduced header, set Confidence to LOW, and warn that freshness cannot be verified — the figures may be current or stale, and the reader should find out which before acting.

Never silently substitute training-data estimates for current prices. When in doubt, ask the user to paste the latest quote.


📋 Data & Sources Header — Open Every Output With It

The first thing in the output is this provenance block, filled in — never left as placeholders. It is the standard documented on the Data & Accuracy page and the first thing result-validator looks for:

Data & Sources
  As of:      <date the figures represent, e.g. 2026-06-30>
  Source:     <primary docs — SEC EDGAR 10-K/10-Q, company IR, FRED, exchange data …>
  Retrieval:  <pasted by user | web/tool retrieval | model memory>
  Confidence: <HIGH | MEDIUM | LOW>
  • Retrieval: model memory must be paired with Confidence: LOW — memory is a placeholder until confirmed against a primary source.
  • Mixed sources: list each with its own as-of date rather than blending them.
  • Sources disagree on the same figure (a closing price, a deal term): prefer the primary source (SEC filing, company release, exchange data) over an aggregator, then the most recent dated one. Show both values with their sources and dates, say which you used and why, and never average them.
  • Data the user pasted is reported as pasted by user; do not upgrade its confidence beyond what the user's own source supports.

For this skill: reproduce the pasted analysis's own header at the top of the explanation. If the analysis has none, write one that says Source: the pasted analysis (no provenance stated), Retrieval: pasted by user, Confidence: LOW — and make the missing header the first thing you teach (see Phase 1).


🌐 Output Language — --lang zh-TW

English is the default. With --lang zh-TW — or when the user writes in Traditional Chinese and asks for a Chinese answer — write the whole analysis in Traditional Chinese (繁體中文, Taiwan usage), keeping it machine-checkable:

  • Translate every heading, narrative paragraph, table header and bullet. Give each metric's English name in brackets the first time it appears — 自由現金流(Free Cash Flow, FCF) — so the reader can match it to the glossary and to English analyses.
  • Keep in English tickers, figures and units, SEC form and item names (10-K Item 7, Form 4), the four Data & Sources field labels (As of, Source, Retrieval, Confidence — the Retrieval value may be written in Chinese, e.g. 使用者提供), and the Investment Signal box and JSON footer exactly as specified — field names and the BULLISH / NEUTRAL / BEARISH, HIGH / MEDIUM / LOW, BUY / HOLD / SELL, STRONG / MODERATE / WEAK values. Tools parse them.
  • Title the invalidation section 論點失效條件(Thesis Invalidation), and close with the disclaimer in Chinese: 免責聲明: 僅供教育用途,非投資建議。
  • Use Taiwan financial terms: 殖利率、本益比、營業利益率、自由現金流、聯準會、選擇權、買權/賣權、除息日、預扣稅.

Overview

Every lesson on the Learning track says "here is the idea." Every skill in the catalog says "here is the number." Nothing sits between them. A beginner who runs stock-eval gets a Piotroski score, an ROIC, a moat rating, and a BULLISH box — and has no way to know which of those actually drove the call, which range counts as good for this sector, or what they should now be able to say about the company in their own words.

This skill is that bridge. It takes any InvestSkill output — a single-skill run, a full-report, a bear-case, a portfolio-review — and explains it the way a patient mentor would: one metric at a time, in plain words, with the good/bad range, the reason it matters for the decision, and the lesson or glossary entry that teaches it properly. Then it turns the reader from passive to active with a short ladder of Socratic questions and the one question every investor should be able to answer: what would change your mind?

It closes the loop the project was built for: run a skill → run the coach → answer its questions → re-run the skill with better inputs.

What it is not. It does not re-analyze, re-score, or second-guess the pasted output. The signal stays exactly as the original skill produced it. If the analysis looks wrong or unsupported, the coach says why that matters and how to check it — and points to result-validator, which is the skill built to audit an analysis. The coach teaches; the validator judges.

Pairs with: every analysis skill (as input), result-validator (when the reader wants a verdict rather than a lesson), and the whole Learning track.


1. Modes

ModeInvocationWhat it does
Explain (default)paste any InvestSkill output, optionally --level beginner | intermediate and --lang zh-TWExplanation card per metric → how the pieces connect → common misreadings → Socratic ladder → "what would change your mind?"
Quiz--quiz <lesson> where lesson ∈ Foundations · Statements · Quality · Valuation · Market · Portfolio · Playbook · Case AMD · Setup (Before Your First Trade) · ETFs · Taxes · Earnings Season · Psychology · Macro & the Fed · Options · Case Pass (When the Answer Is No)Five questions on one Learning lesson (recall · calculation · judgment), answers hidden under an Answers heading, one line on what to re-read

Flags:

  • --level beginner (default) — no term is used before it is defined in one sentence; analogies allowed; at most one formula per card.
  • --level intermediate — assumes the Foundations and Statements lessons; skips definitions of revenue, margin, EPS, P/E, free cash flow; spends the words on interpretation and sector context instead.
  • --lang zh-TW — the whole output in Traditional Chinese; keep every metric's English name in brackets after the Chinese term the first time it appears, e.g. 自由現金流(Free Cash Flow, FCF), so the reader can match it to the English analysis and glossary.

2. Inputs

InputRequiredDefault if unstated
A pasted InvestSkill output (any skill, any length)✅ for Explainask — the coach cannot explain what it has not seen
Lesson name✅ for Quiz—
--leveloptionalbeginner
--langoptionalEnglish
What the reader already knows / is confused byoptionalinfer from the level; if the user names a metric, start there

If the user pastes a question about a metric without an analysis ("what is ROIC?"), answer it as a single explanation card and point to the lesson — do not invent an analysis to wrap it in.


3. Framework — Explain Mode

Work top to bottom through the pasted analysis. Keep the reader's attention on the three or four metrics that actually moved the signal; list the rest briefly. A 600-line full-report does not need 600 lines of explanation.

Phase 1 — Read the contract first

Before any metric, check the three things every InvestSkill output must carry, and teach each one the reader is missing:

Contract pieceIf presentIf missing — teach this
Data & Sources headerReproduce it; explain in one sentence what Retrieval and Confidence tell the reader about how much to trust every number below"Without this header you cannot tell whether the P/E came from a filing or from the model's memory. Treat every figure as a placeholder until it is sourced. result-validator will cap Data Quality for exactly this reason."
Thesis Invalidation sectionSave it — it becomes the material for Phase 5"An analysis that cannot say what would make it wrong is an opinion, not a thesis. Ask the original skill to add it, or write your own three triggers."
Investment Signal blockRead Signal · Score · Confidence · Action; check the score sits in the band its signal claims (8–10 strongly bullish, 6–7.9 moderately bullish, 4–5.9 neutral, 2–3.9 moderately bearish, 0–1.9 strongly bearish)"No signal block means the analysis never committed to a conclusion — it described the company but did not decide anything."

Phase 2 — Explanation card, one per metric

Use this card for every metric you explain. Beginner level fills every row; intermediate may drop Plain words when the term is basic.

▸ <Metric name>  ·  value in this analysis: <as stated>
  Plain words     one sentence a non-finance friend would follow
  Why it matters  what decision it informs, and what happens to the thesis if it is wrong
  Good / bad      the general range, then the range for THIS sector or business model
  In this case    what the stated value says about this company, in one line
  Learn it        <Lesson name> → https://yennanliu.github.io/InvestSkill/learning-<slug>.html
                  Glossary → https://yennanliu.github.io/InvestSkill/glossary.html

Rules for the card:

  • Ranges are sector-relative. A 30% operating margin is superb for a retailer and ordinary for enterprise software; a P/E of 12 is cheap for a compounder and fair for a utility. Always give both the generic range and the sector range, and say which one applies.
  • One formula at most, and only if it aids understanding (e.g. FCF = operating cash flow − capex). Show it with the analysis's own numbers when they are present.
  • Never restate the analysis's judgment as your own. "The analysis rates the moat WIDE because…" — not "the moat is wide."
  • Order the cards by influence on the signal, not by their order in the pasted text.

Mapping metrics to lessons (use these slugs):

Metric familyLessonSlug
Revenue, margins, EPS, cash flow, balance-sheet itemsStatementslearning-statements
ROIC, ROE, Piotroski F-Score, accruals, moat, capital allocationQualitylearning-quality
P/E, P/S, EV/EBITDA, DCF, WACC, terminal growth, fair-value rangeValuationlearning-valuation
Beta, moving averages, RSI, MACD, short interest, implied volatility, insider / 13F flowsMarketlearning-market
Position size, concentration, correlation, drawdown, VaR, rebalancingPortfoliolearning-portfolio
Signal block, thesis, invalidation triggers, catalysts, "when to sell"Playbooklearning-playbook
The whole loop on one real companyCase AMDlearning-case-amd
Compounding, risk vs. return, what a stock isFoundationslearning-foundations
Account types, order types, settlement (T+1), the IPS, risk capacity vs. toleranceSetuplearning-setup
Expense ratio, tracking difference, ETF liquidity, holdings overlap, leveraged/inverse structureETFslearning-etfs
Short- vs. long-term gains, wash sale, qualified dividends, lots, harvesting, W-8BEN, withholdingTaxeslearning-taxes
Consensus, guidance, beat rate, implied move, "beat and drop", post-earnings driftEarnings Seasonlearning-earnings
Biases, process vs. outcome, pre-mortem, journal, sell rules, trade reviewPsychologylearning-psychology
Policy rate, yield curve, CPI/PCE, PMI, LEI, Sahm rule, cycle phase, sector rotationMacro & the Fedlearning-macro
Calls, puts, the Greeks, IV and IV rank, covered calls, cash-secured and protective putsOptionslearning-options
A thesis that ends in no — value traps, no-trade setupsCase Passlearning-case-pass

Phase 3 — How the pieces connect

One paragraph, no more than six sentences: which two or three metrics carried the signal, which ones argued against it, and how the original skill weighed them. Name the trade-off explicitly ("high quality, but the price already assumes it"). This is the paragraph the reader should be able to repeat from memory tomorrow.

Phase 4 — Common misreadings

List the three to five errors beginners make with this kind of output. Draw from this bank and add skill-specific ones:

  • Treating a high score or Action: BUY as an order to buy today, rather than as one input to a position decision (position-ladder, portfolio-review decide size and timing).
  • Reading the signal box without reading the Data & Sources header above it — the box inherits the header's confidence.
  • Confusing a low P/E with cheapness (a low multiple on falling earnings is a value trap — see the PFE run in the Cookbook) or a high P/E with expensiveness (growth can justify it; the reverse-DCF in stock-valuation says whether it does).
  • Reading Confidence: HIGH as "high probability the stock goes up" — it means the data was strong and the signals agreed, not that the outcome is likely.
  • Taking a bear-case output as a balanced verdict — it is deliberately one-sided by design.
  • Treating a DCF fair value as a fact rather than the output of stated assumptions (change the terminal growth by 1% and watch the value move).
  • Anchoring on the average cost or the entry price — the market does not know what you paid.
  • Reading Horizon: LONG-TERM as permission to stop checking — the Thesis Invalidation section says when to look again.

Phase 5 — The Socratic ladder

Three to five questions, each harder than the last, each answerable from the pasted analysis plus the cards above. Do not answer them; end the ladder with the standing question.

RungTypeExample shape
1Recall"Which single metric in this analysis most supports the BULLISH call, and what is its value?"
2Interpretation"The analysis says gross margin is 75% and expanding. What does an expanding margin tell you that a high margin alone does not?"
3Connection"If revenue growth slowed to 10% next year but margins held, which of the analysis's conclusions would survive and which would not?"
4Judgment"The valuation section says the price assumes 25% growth for five years. Is that plausible for a company this size? What would you need to see to believe it?"
5Transfer"Take the same three metrics and apply them to a competitor you know. Does the comparison change how you read this company?"
—Standing question"What would change your mind?" — answer with the analysis's own Thesis Invalidation triggers; if it has none, write three.

4. Framework — Quiz Mode (--quiz <lesson>)

Five questions on one Learning lesson, in this fixed mix:

#TypeDesign rule
1RecallA definition or a "which statement is true" — answerable straight from the lesson
2RecallA second core term, ideally the one beginners confuse with the first
3CalculationGive a tiny, fully specified data set (three to five numbers) and ask for one computed value — e.g. "Revenue $200M, COGS $120M, opex $50M: what is the operating margin?" Show the worked solution in the answers
4JudgmentTwo companies described in two lines each; ask which the lesson's framework prefers and why
5Judgment / transferA situation where the naive reading and the correct reading differ (the value-trap pattern, the buyback that hides dilution, the beta that understates risk)

Format:

## Quiz — <Lesson name>  (5 questions · --level <level>)

1. …
2. …
3. … (data: …)
4. …
5. …

## Answers
1. <answer> — <one-line why>
2. …
3. <worked calculation> = <result>
4. …
5. …

**Re-read:** <Lesson name> → <section or concept>, and the Glossary entry for <term>.

Never pull questions from outside the named lesson; the point is to confirm the lesson landed, not to test breadth. In --lang zh-TW write questions and answers in Traditional Chinese with English metric names in brackets.


5. Output Format

Explain mode

  1. Data & Sources header (the pasted analysis's own, or the LOW-confidence substitute)
  2. One-line orientation — which skill produced this, what question it answers, and its signal as stated
  3. Contract check (Phase 1) — only the pieces that are missing get a paragraph; present pieces get one line each
  4. Explanation cards (Phase 2), ordered by influence on the signal — three to six full cards, the remaining metrics in a compact table (metric · value · one-line meaning · lesson)
  5. How the pieces connect (Phase 3)
  6. Common misreadings (Phase 4)
  7. Socratic ladder (Phase 5), ending with What would change your mind?
  8. Next step — one line: which lesson to read next, and which skill to run to test the reader's new question
  9. Thesis Invalidation (the reader's version — below) and the Investment Signal block, mirroring the pasted analysis's signal unchanged, with the Learning Coach note

Quiz mode — the quiz block from §4, then the Thesis Invalidation and signal sections below (signal fields may be n/a — a quiz has no analysis to mirror; say so in the note).


Example

User: [pastes a stock-eval output for a semiconductor company: Piotroski 8/9, ROIC 34%,
      gross margin 75% expanding, forward P/E 38, moat WIDE, signal BULLISH 7.8/10,
      Confidence HIGH, Data & Sources header present, Thesis Invalidation present]
      Explain this to me — I'm new to this.

The assistant reproduces the Data & Sources header, notes all three contract pieces are
present, then writes five cards ordered by influence: ROIC (plain words: "for every $100 the
company invests in itself, it earns $34 a year"; good/bad: >15% strong generally, >25% elite
for semis; Learn it: Quality lesson), gross margin and its trend, the Piotroski F-Score
(what 8 of 9 means and which test it failed), forward P/E of 38 (generic: high; sector: the
upper half for growth semis; "the price already assumes the growth continues"), and the WIDE
moat rating (which of the five moat sources the analysis relied on). The connecting paragraph
says quality metrics carried the call and valuation argued against it, and that the 7.8 is
"quality strong enough to pay up for, but not cheap." Misreadings: 7.8 is not a buy order;
HIGH confidence describes the data, not the odds; P/E 38 is not automatically expensive.
The ladder ends: "What would change your mind?" — and quotes the analysis's own three
invalidation triggers (margin below 70%, growth below 40%, a top customer moving in-house).
Next step: read the Valuation lesson, then run stock-valuation to see what growth the
price implies. The signal block repeats BULLISH · 7.8 · HIGH unchanged.

User: --quiz Valuation --level beginner

The assistant writes five questions on the Valuation lesson — two definitions (intrinsic
value vs. price; what a discount rate is), one calculation (FCF $500M, 20 shares of 50M...
"what is FCF per share, and at a 5% FCF yield what price does that imply?" → $10/share,
$200), one two-company judgment (same P/E, different growth — which is cheaper?), and one
trap (a low P/E on earnings that just halved). Answers under the Answers heading with the
calculation worked out. Re-read: Valuation → "price vs. value", Glossary → FCF yield.

Notes

  • Explain, don't re-analyze. If the reader wants to know whether the analysis is right, send them to result-validator; if they want the opposite view, to bear-case. The coach's value is that it never changes the answer — it changes what the reader understands about it.
  • Match the level honestly. A beginner who receives an intermediate explanation learns nothing; an intermediate reader who receives definitions of revenue stops reading. When unsure, ask one question about what they already know before writing.
  • Sector ranges are the hardest part and the most valuable. Always state which sector's norms you are using and why; when the pasted analysis names peers, use them.
  • Quote the analysis, cite the lesson. Every card should point at where in the pasted text the number came from and where on the site the concept is taught — the reader should be able to check both.
  • Traditional Chinese output follows the Learning track's own zh-TW pages (learning-zh-tw.html, glossary-zh-tw.html); link those instead of the English pages when --lang zh-TW is set.
  • Long inputs: a full-report may contain fifteen modules. Explain the composite and the three modules with the largest weight fully; summarize the rest in the compact table and offer to expand any one on request.

Thesis Invalidation

This skill explains a reading rather than making a call, so what can be invalidated is the learner's understanding. After the explanation, state when it stops being valid:

The explanation no longer applies if:

  • The analysis is re-run and any of the three signal-driving metrics changes band (the cards were written for these values — re-run the coach on the new output rather than re-reading the old one)
  • The Data & Sources header dates the figures more than 90 days back, or a newer filing has landed since (the meaning of a metric does not age; its value does)
  • The pasted analysis is later shown by result-validator to have a data or arithmetic error — an explanation of a wrong number is still wrong, however clear
  • The reader has moved from beginner to intermediate — the beginner cards now under-explain interpretation and over-explain definitions; re-run at the higher level

The learner's own reading is invalidated if they cannot:

  • Name the two or three metrics that drove the signal, without looking
  • State the good / bad range for the key metric in this sector
  • Answer "what would change your mind?" with a specific, observable trigger

Re-run this coach when:

  • The underlying analysis is re-run (next earnings, price move ±15%, material news)
  • The reader finishes the lesson the cards pointed to — a second pass should feel different
  • The reader runs a different skill on the same company (the connections paragraph changes)
  • 60 days have elapsed and the reader wants to check what stuck (use --quiz)

Standard Signal Output

This skill does not produce a signal of its own. The block below mirrors the pasted analysis's Investment Signal exactly — same Signal, Confidence, Horizon, Score, Action, and Conviction — so the reader sees the conclusion they were just taught to read, unchanged. If the pasted material has no signal block (or the run is a --quiz), fill every field with n/a and say so in the note.

All analysis concludes with this standardized block:

In the output, the box below is followed immediately by this machine-readable footer — the same values as the box, filled in (no placeholders, no comments), so tools can read the result without parsing the box:

{"investskill": "1", "skill": "learning-coach", "ticker": "<TICKER, or null for a portfolio / market-wide view>", "as_of": "<YYYY-MM-DD from Data & Sources>", "signal": "<BULLISH | NEUTRAL | BEARISH>", "confidence": "<HIGH | MEDIUM | LOW>", "horizon": "<SHORT-TERM | MEDIUM-TERM | LONG-TERM>", "score": 0.0, "action": "<BUY | HOLD | SELL>", "conviction": "<STRONG | MODERATE | WEAK>", "retrieval": "<pasted by user | web/tool retrieval | model memory>", "sources": ["<each primary source, dated>"]}
╔══════════════════════════════════════════════╗
║              INVESTMENT SIGNAL               ║
╠══════════════════════════════════════════════╣
║ Signal:      BULLISH / NEUTRAL / BEARISH     ║
║ Confidence:  HIGH / MEDIUM / LOW             ║
║ Horizon:     SHORT / MEDIUM / LONG-TERM      ║
║ Score:       X.X / 10                        ║
╠══════════════════════════════════════════════╣
║ Action:      BUY / HOLD / SELL               ║
║ Conviction:  STRONG / MODERATE / WEAK        ║
╚══════════════════════════════════════════════╝

Score Guide: 8.0–10.0 Strongly Bullish | 6.0–7.9 Moderately Bullish | 4.0–5.9 Neutral | 2.0–3.9 Moderately Bearish | 0.0–1.9 Strongly Bearish Confidence: HIGH (strong data, clear signals) | MEDIUM (mixed signals) | LOW (limited data, conflicting signals) Horizon: SHORT-TERM (1 week–3 months) | MEDIUM-TERM (3 months–1 year) | LONG-TERM (1+ years)

Learning Coach note: (one line) — "After this explanation you should be able to say, in your own words, why the analysis reached <Signal> and which <metric> would have to change to reverse it." The box above is the analysed skill's conclusion, reproduced without alteration; the coach's contribution is the understanding, not the verdict.

Disclaimer: Educational content only. Not financial advice. This skill explains an analysis; it does not endorse, verify, or update its conclusion.

Options Analysis

⚠️ Data Verification — Do This Before Any Analysis

Before running any analysis, always retrieve the latest market data for the ticker:

  1. Fetch current price — use web search or ask the user for the live price, 52-week range, and market cap. Never assume a price from training data.
  2. Confirm key figures — the option chain as of a timestamp (strikes, expiries, bid/ask, volume, open interest), implied volatility and IV rank, the underlying price, the next earnings and ex-dividend dates, and the risk-free rate.
  3. State your data source — fill in the Data & Sources header (next section) so the origin, as-of date, retrieval path, and confidence of every figure are explicit at the top of the output.
  4. Flag stale data explicitly — if live data is unavailable, display this warning before proceeding:

⚠️ Live data unavailable. The following analysis uses training-data estimates which may be significantly out of date. Verify all prices and metrics before making any decisions.

Never silently substitute training-data estimates for current prices. When in doubt, ask the user to paste the latest quote.


📋 Data & Sources Header — Open Every Output With It

The first thing in the output is this provenance block, filled in — never left as placeholders. It is the standard documented on the Data & Accuracy page and the first thing result-validator looks for:

Data & Sources
  As of:      <date the figures represent, e.g. 2026-06-30>
  Source:     <primary docs — SEC EDGAR 10-K/10-Q, company IR, FRED, exchange data …>
  Retrieval:  <pasted by user | web/tool retrieval | model memory>
  Confidence: <HIGH | MEDIUM | LOW>
  • Retrieval: model memory must be paired with Confidence: LOW — memory is a placeholder until confirmed against a primary source.
  • Mixed sources: list each with its own as-of date rather than blending them.
  • Sources disagree on the same figure (a closing price, a deal term): prefer the primary source (SEC filing, company release, exchange data) over an aggregator, then the most recent dated one. Show both values with their sources and dates, say which you used and why, and never average them.
  • Data the user pasted is reported as pasted by user; do not upgrade its confidence beyond what the user's own source supports.

🌐 Output Language — --lang zh-TW

English is the default. With --lang zh-TW — or when the user writes in Traditional Chinese and asks for a Chinese answer — write the whole analysis in Traditional Chinese (繁體中文, Taiwan usage), keeping it machine-checkable:

  • Translate every heading, narrative paragraph, table header and bullet. Give each metric's English name in brackets the first time it appears — 自由現金流(Free Cash Flow, FCF) — so the reader can match it to the glossary and to English analyses.
  • Keep in English tickers, figures and units, SEC form and item names (10-K Item 7, Form 4), the four Data & Sources field labels (As of, Source, Retrieval, Confidence — the Retrieval value may be written in Chinese, e.g. 使用者提供), and the Investment Signal box and JSON footer exactly as specified — field names and the BULLISH / NEUTRAL / BEARISH, HIGH / MEDIUM / LOW, BUY / HOLD / SELL, STRONG / MODERATE / WEAK values. Tools parse them.
  • Title the invalidation section 論點失效條件(Thesis Invalidation), and close with the disclaimer in Chinese: 免責聲明: 僅供教育用途,非投資建議。
  • Use Taiwan financial terms: 殖利率、本益比、營業利益率、自由現金流、聯準會、選擇權、買權/賣權、除息日、預扣稅.

Comprehensive analysis of options Greeks, implied volatility dynamics, strategy selection, and risk/reward frameworks for US-listed equity options. Designed to help identify optimal strategies based on market outlook, volatility environment, and underlying stock context.

Risk Disclaimer: Options involve significant leverage and can result in the loss of the entire premium paid or, for short options, losses substantially exceeding the initial credit received. This analysis is for educational and informational purposes only and does not constitute financial advice.

Analysis Framework

1. Options Basics and Context

Core Terminology

  • Strike price: The price at which the option holder can buy (call) or sell (put) the underlying stock
  • Expiration date: The date the option contract expires; American-style options can be exercised at any time before expiration
  • Premium: The market price of the option contract (cost to buyer, credit received by seller)
  • Contract: Represents 100 shares of the underlying stock
  • ITM (In-the-Money): Call with strike < stock price; Put with strike > stock price — has intrinsic value
  • ATM (At-the-Money): Strike approximately equal to current stock price
  • OTM (Out-of-the-Money): Call with strike > stock price; Put with strike < stock price — only time value, no intrinsic value

Intrinsic Value vs. Time Value

  • Premium = Intrinsic Value + Time Value (Extrinsic Value)
  • Intrinsic value: how far ITM the option is (cannot be negative)
  • Time value: premium above intrinsic value — represents probability of further price movement and time remaining

When to Use Options vs. Stock

  • Leverage: Control 100 shares with a fraction of the capital (calls vs. outright long)
  • Hedging: Protect existing portfolio positions (protective puts, collars)
  • Income generation: Sell covered calls or cash-secured puts to generate premium income
  • Volatility plays: Profit from large moves (straddles) or stability (iron condors) regardless of direction
  • Defined risk: Spreads cap maximum loss to the debit paid

2. The Greeks Analysis

Delta (Δ) — Directional Sensitivity

Definition: Change in option price for a $1 move in the underlying stock price.

Option Type      Delta Range      Moneyness Context
Call (long)      0 to +1.0        Deep OTM ≈ 0.05, ATM ≈ 0.50, Deep ITM ≈ 0.99
Put (long)       -1.0 to 0        Deep OTM ≈ -0.05, ATM ≈ -0.50, Deep ITM ≈ -0.99
  • Delta as probability proxy: A 70-delta option approximates a 70% probability of expiring ITM (rough heuristic, not precise)
  • Position delta: Sum of all delta exposure across a portfolio = total directional equivalent shares
  • Directional traders: select delta matching conviction (high delta = high directional exposure, lower leverage; low delta = cheaper, more speculative)
  • Delta hedging: offset option delta with opposite stock position to create delta-neutral portfolio

Gamma (Γ) — Rate of Delta Change

Definition: Rate of change of delta for a $1 move in the underlying.

  • Gamma is highest for ATM options and increases dramatically as expiration approaches
  • Long gamma: Long options positions (long calls or puts) — benefits from large moves in either direction; gamma accelerates gains as stock moves through strike
  • Short gamma: Short options positions — benefits from stability and small moves; faces accelerating losses on large moves
  • Gamma risk at expiration: ATM options near expiry have very high gamma — pin risk and rapid delta shifts
  • Gamma scalping: Dynamically delta-hedging a long gamma position to profit from realized volatility exceeding implied volatility
  • Weekly options have the highest gamma, making them both powerful and dangerous near expiry

Theta (Θ) — Time Decay

Definition: Daily erosion of an option's value due to the passage of time (always negative for long options, positive for short options).

  • Theta represents the "rent" paid to maintain a long options position
  • Time decay is not linear — it accelerates as expiration approaches
DTE Range        Theta Decay Speed
>90 days         Slow — manageable for long holders
60-90 days       Moderate — beginning to accelerate
30-60 days       Elevated — meaningful daily decay
15-30 days       Fast — significant daily cost
7-15 days        Rapid — theta dominates P&L
<7 days          Extreme — ATM options decay very rapidly
  • Theta / premium: The daily decay as a percentage of the option's premium
  • Short options strategies (iron condors, covered calls) benefit from theta; structure positions so that time is your ally
  • Rule of thumb: avoid holding long OTM options into the final 21 days unless near a catalyst

Vega (ν) — Volatility Sensitivity

Definition: Change in option price for a 1 percentage point change in implied volatility (IV).

  • Long options = long vega: Long calls and long puts both benefit from IV expansion
  • Short options = short vega: Short premium strategies benefit from IV contraction (volatility crush)
  • Vega is largest for ATM options with 30-60 days to expiration
  • Earnings and binary events dramatically increase vega exposure — IV expands before event, then collapses after (volatility crush)
IV Change Impact on Long Option:
+5% IV increase on $5.00 option with vega 0.20 = +$1.00 gain
-5% IV decrease (post-earnings crush) = -$1.00 loss
  • Vega term structure: Long-dated options have higher absolute vega — more sensitive to IV changes
  • Managing vega: use spreads to reduce net vega exposure (long spread = smaller net vega than long naked option)

Rho (ρ) — Interest Rate Sensitivity

Definition: Change in option price for a 1 percentage point change in the risk-free interest rate.

  • Calls have positive rho: benefit from interest rate increases (higher rates increase call values)
  • Puts have negative rho: decrease in value as interest rates rise
  • Rho is typically the least impactful Greek for short-dated options
  • More relevant for LEAPS (long-dated options with 1-2 year expirations) in a higher interest rate environment
  • In a high-rate environment (rates >4%): rho becomes a meaningful consideration for 6-12 month options strategies

3. Implied Volatility (IV) Analysis

IV vs. Historical Volatility (HV)

  • IV (Implied Volatility): Forward-looking volatility derived from current option prices — reflects market's expectation of future price swings
  • HV (Historical Volatility): Realized past volatility measured over 20, 30, or 60 days
  • IV > HV: Options are expensive relative to recent realized movement — favor selling premium (credit strategies)
  • IV < HV: Options are cheap relative to recent realized movement — favor buying premium (debit strategies)
  • IV/HV ratio: >1.2 = meaningfully overpriced, <0.8 = meaningfully underpriced

IV Rank (IVR)

Formula: IVR = (Current IV - 52-Week Low IV) / (52-Week High IV - 52-Week Low IV) × 100

Example: Current IV 35%, 52-wk Low 20%, 52-wk High 60%
IVR = (35 - 20) / (60 - 20) × 100 = 37.5%

IVR Interpretation:
>70%    Very High — IV near 52-week highs, strongly favor selling premium
50-70%  High — elevated IV, lean toward selling premium
30-50%  Moderate — mixed, direction-dependent strategies
<30%    Low — IV near 52-week lows, favor buying premium or debit spreads
<15%    Very Low — cheapest options of the year, consider long volatility

IV Percentile (IVP)

  • Definition: Percentage of trading days in the past 252 days where IV was lower than the current reading
  • IVP 80% = IV is higher today than on 80% of trading days in the past year
  • More statistically robust than IVR because IVR can be skewed by a single outlier high or low
  • IVP > 50%: favor selling premium | IVP < 30%: favor buying premium

IV Term Structure

  • Contango (normal): Near-term IV < far-term IV — calm near-term environment, uncertainty longer-dated
  • Backwardation (inverted): Near-term IV > far-term IV — acute fear or event in immediate future
  • Steep backwardation: market pricing major near-term risk (earnings miss, regulatory event, macro shock)
  • Calendar spread opportunity: sell high near-term IV, buy lower far-term IV (take advantage of backwardation)

IV Skew Analysis

  • Put skew: OTM put IV > OTM call IV — standard condition in equity markets (portfolio protection demand)
  • Steep negative skew: Market pricing significant downside protection — fear elevated, consider selling OTM puts
  • Skew reversal (call skew): OTM call IV > OTM put IV — squeeze expectation, binary upside event anticipated
  • Risk reversal metric: 25-delta put IV minus 25-delta call IV. Higher value = steeper put skew.
  • Skew changes directionally: steepening skew = increasing fear, flattening skew = reducing downside concern

Volatility Crush (Post-Event IV Collapse)

  • IV systematically rises into known binary events: earnings, FDA decisions, investor days, FOMC
  • IV collapses immediately after the event as uncertainty resolves — regardless of the price move direction
  • Average IV crush on earnings: 30-50% drop in IV on the day following announcement
  • Implication: long premium strategies must have a move that exceeds the expected move to profit; short premium strategies profit even if stock moves, as long as within expected range

4. Strategy Selector

Select the optimal strategy based on market outlook and IV environment:

Bullish Strategies

StrategyWhen to UseIV PreferenceMax ProfitMax Loss
Buy CallStrongly bullish, conviction directional betLow IV (IVR <30%)UnlimitedPremium paid
Bull Call SpreadModerately bullish, reduce cost, cap profitAny IVStrike spread minus debitDebit paid
Cash-Secured PutBullish/neutral, willing to own stock at lower priceHigh IV (IVR >50%)Premium collectedStrike minus premium
Covered CallBullish/neutral, income on existing positionHigh IV (IVR >50%)Premium + upside to strikeStock cost minus premium

Bearish Strategies

StrategyWhen to UseIV PreferenceMax ProfitMax Loss
Buy PutStrongly bearish, protection or directional betLow IV (IVR <30%)Strike minus premiumPremium paid
Bear Put SpreadModerately bearish, reduce cost, define riskAny IVStrike spread minus debitDebit paid
Bear Call SpreadBearish/neutral, sell premium with defined max lossHigh IV (IVR >50%)Credit receivedStrike spread minus credit

Neutral and Volatility Strategies

StrategyWhen to UseIV PreferenceMax ProfitMax Loss
Iron CondorNeutral price, range-boundHigh IV (IVR >50%)Full creditWing spread minus credit
Iron ButterflyNeutral, max decay at single price pointHigh IV (IVR >50%)Full creditWing spread minus credit
Short StraddleNeutral, sell both call and put ATMVery High IVFull creditUnlimited (uncapped)
Long StraddleExpecting large move, unknown directionLow IV (IVR <20%)UnlimitedBoth premiums paid
Long StrangleLarge move expected, OTM for lower costLow IV (IVR <20%)UnlimitedBoth premiums paid

Strategy Selection Matrix

                        IV High (IVR >50%)       IV Low (IVR <20%)
─────────────────────────────────────────────────────────────────────
Strongly Bullish:       Bull Call Spread          Buy Call
Mildly Bullish:         Cash-Secured Put          Bull Call Spread
Neutral:                Iron Condor               Long Straddle / Strangle
Mildly Bearish:         Bear Call Spread          Bear Put Spread
Strongly Bearish:       Bear Put Spread           Buy Put
High Volatility:        Short Straddle            Long Straddle
Low Conviction:         Iron Condor               Calendar Spread
─────────────────────────────────────────────────────────────────────

5. Earnings Play Analysis

IV Crush Dynamics

  • IV rises steadily in the 2-4 weeks before earnings as uncertainty increases
  • On the day after earnings announcement, IV collapses 30-50% (volatility crush)
  • This crush destroys long premium value even if the stock moves significantly
  • Short premium strategies capture this crush but face the binary event risk

Expected Move Calculation

Expected Move (EM) = ATM Straddle Price / Stock Price × 100

Example: Stock at $100, ATM straddle costs $7.00
Expected Move = 7.0 / 100 × 100 = 7.0%
Market implying 68% probability stock stays within +/-$7 of current price

Historical Earnings Move Analysis

  • Compare actual historical earnings moves vs. expected move at time of earnings
  • If stock regularly moves ≥ 1.5x the expected move: favor buying straddle/strangle (breakeven likely exceeded)
  • If stock regularly moves < 0.7x the expected move: favor selling premium (iron condor, short strangle)
  • Look at last 8 quarterly earnings reactions for statistically meaningful sample

Earnings Straddle Setup

  • Buy ATM call + ATM put with same strike and expiration just before earnings
  • Profit if stock moves more than combined cost of both options (breakeven points = strike ± total premium)
  • Risk: theta decay into earnings + IV crush if move is below expected move
  • Optimal entry: 7-14 days before earnings (balances theta risk vs. IV run-up capture)

Defined Risk Earnings Structures

  • Long strangle: OTM call + OTM put, cheaper than straddle, requires larger move to profit
  • Broken wing butterfly: Define maximum risk, profit if stock pins near current level, smaller debit
  • Earnings iron condor: Sell wings around expected move, profit if stock stays within range

6. Risk Management

Maximum Loss Scenarios by Strategy

  • Long call / Long put: Max loss = premium paid (100% of investment)
  • Bull/Bear spread: Max loss = net debit paid
  • Iron condor: Max loss = wing width minus total credit received
  • Short straddle / strangle: Max loss = theoretically unlimited (capped by stock going to zero for puts)
  • Covered call: Max loss = stock cost basis minus premium received (stock declining to zero)

Position Sizing

  • Long options premium risk: limit to 2-5% of portfolio per trade (entire premium can be lost)
  • Short premium, defined-risk (credit spreads, iron condors, iron butterflies): max loss = spread width − credit; size so that max loss is 1-3% of portfolio
  • Short premium, undefined-risk (short straddles, short strangles, naked puts/calls): loss is not capped by a spread width — size on a stress loss instead: assume a 2-standard-deviation move (or the stock's worst historical earnings gap, whichever is larger) and keep that loss to 1-3% of portfolio; keep buying-power reduction under 20% of the account; and never carry an undefined-risk position through earnings without a pre-set exit level
  • Never let a single options position represent more than 10% of total portfolio market value
  • For speculative OTM long options: size even smaller (0.5-1%) given low probability of profit

Delta Hedging for Complex Positions

  • Monitor portfolio delta daily for active options books
  • Adjust underlying stock position to maintain target delta exposure (delta-neutral if desired)
  • Frequency of hedging: daily for near-expiry positions, weekly for longer-dated positions
  • Hedging costs: transaction costs erode gamma scalping profits; only hedge on meaningful delta drift

Rolling Positions

  • When to roll: option approaching expiration with significant unrealized gain or at risk of assignment
  • Rolling out: buy to close current expiration, sell to open further expiration (extends duration)
  • Rolling out and up/down: change both expiration and strike
  • Roll mechanics: attempt to collect additional credit (for short options) or minimal debit

Stop-Loss Guidelines

  • Long options (debit): consider closing when position loses 50% of premium paid
  • Short options (credit): consider closing when position reaches 2x the credit received (200% of max profit in losses)
  • Hard rule: never let a defined-risk spread reach maximum loss without reassessment

Common Mistakes to Avoid

  • Holding long OTM options to expiration hoping for recovery (usually expires worthless)
  • Selling naked short puts or calls without understanding assignment risk
  • Ignoring bid/ask spread in illiquid options (wide spread = immediate cost)
  • Over-leveraging via too many contracts on speculative OTM options
  • Forgetting earnings/ex-dividend dates that create event risk on existing positions

7. Options Chain Reading Guide

Key Chain Columns

  • Bid / Ask: Market-maker buy price and sell price — execute near midpoint for better fills
  • Last trade price: Often stale; use bid/ask midpoint for current value
  • Volume: Contracts traded today — high volume = active interest, easier to fill
  • Open Interest (OI): Total outstanding contracts — high OI = liquidity and tight spreads
  • IV per strike: Varies by strike (shows skew visually across the chain)
  • Delta, Gamma, Theta, Vega: Displayed per-contract for each strike

Bid/Ask Spread Importance

  • Wide bid/ask spread = illiquid options = avoid or expect significant slippage
  • Good liquidity threshold: bid/ask spread < 10% of option midprice
  • Actively traded tickers (AAPL, SPY, QQQ): penny-wide spreads on ATM options
  • Small-cap or low-volume stocks: may have $0.50+ spreads that erode profitability

Open Interest Accumulation as Price Levels

  • Strikes with very high OI often act as price magnets (max pain theory) near expiration
  • Institutional hedging clusters at round number strikes (every $5 or $10 increments)
  • Dealer gamma exposure at high-OI strikes: dealers hedge dynamically, creating mechanical buying/selling pressure
  • Max pain price: strike at which maximum total options value expires worthless

Volume vs. OI Interpretation

  • Volume > OI: New positions being opened (not just rolls of existing positions)
  • Volume << OI: Trading is closing or rolling existing positions
  • Large single print on low-OI strike: new speculative or informed position (unusual activity alert)
  • Volume spike with no news: potential informed positioning ahead of catalyst

8. Data Sources

Primary Trading Platforms with Full Greeks

  • Thinkorswim (TD Ameritrade / Schwab): Comprehensive options chain, Greeks, IV analysis, probability cones
  • Interactive Brokers (IBKR): Options analytics, volatility trader tools, probability of expiring
  • Tastytrade: Options-specific platform, IV rank/percentile, theta decay tracking

Free Analytics Tools

  • Market Chameleon: IV rank, IV percentile, term structure, earnings IV history — all free
  • Barchart Options: Options flow, unusual activity, Greeks by strike (free tier available)
  • Options Profit Calculator (optionsprofitcalculator.com): Visual P&L diagrams for strategies (free)
  • Power E*TRADE: Options analytics and strategy builder

Options Flow and Unusual Activity

  • Unusual Whales: Options flow tracking, dark pool data, unusual activity scanner
  • Flowalgo: Real-time unusual options activity alerts (premium)
  • Cheddar Flow: Institutional options flow visualization

Reference Data

  • Options Clearing Corporation (OCC): Volume and OI data, contract specifications
  • CBOE: VIX data, volatility indices, SPX/SPY options data

Input Formats

Format 1: Greeks Analysis

User: /options-analysis AAPL --greeks

Claude analyzes ATM and key strikes across near-term expirations, summarizes Greeks profile

Format 2: Strategy Recommendation

User: /options-analysis TSLA --strategy bullish

Claude evaluates current IV environment, selects optimal bullish strategies, provides specific setups

Format 3: Earnings Play Analysis

User: /options-analysis NVDA --earnings

Claude computes expected move, reviews historical earnings reactions, recommends earnings strategy

Format 4: IV Analysis

User: /options-analysis SPY --iv

Claude provides full IV analysis: IVR, IVP, term structure, skew, HV comparison

Output

Provide a comprehensive options analysis report with the following sections:

1. Options Overview

Underlying:              [TICKER]
Current Price:           $XXX.XX
ATM Implied Volatility:  XX.X%
IV Rank (IVR):           XX%    (52-wk Range: XX% - XX%)
IV Percentile (IVP):     XX%
Historical Volatility:   XX.X%  (30-day realized)
IV vs. HV:               X.Xx   (Overpriced / Underpriced)
Next Earnings:           [Date] (XX days)

2. Greeks Summary (ATM Nearest Expiry)

Strike:       $XXX       Expiry: [Date]    DTE: XX
─────────────────────────────────────────
Call Greeks:  Delta +X.XX | Gamma X.XX | Theta -$X.XX/day | Vega $X.XX/1%IV
Put Greeks:   Delta -X.XX | Gamma X.XX | Theta -$X.XX/day | Vega $X.XX/1%IV
─────────────────────────────────────────
Expected Move (±1σ to expiry): $X.XX (X.X%)

3. IV Analysis Summary

  • IV vs. HV assessment (are options cheap or expensive?)
  • IVR/IVP interpretation with action bias
  • Term structure shape (contango vs. backwardation)
  • Skew assessment and directional implication
  • If near earnings: IV run-up remaining and expected crush magnitude

4. Recommended Strategies (Top 2-3)

For each recommended strategy:

  • Strategy name and structure (which strikes, which expiry)
  • Rationale (why this strategy fits current IV and outlook)
  • Max profit / Max loss
  • Breakeven price(s)
  • Probability of profit (from delta approximation)
  • Ideal outcome scenario

5. Key Levels for Options

  • Strike clustering (high OI strikes = market-recognized levels)
  • Support / resistance relevant to spread placement
  • Max pain price for nearest monthly expiry

6. Entry and Exit Guidelines

  • Suggested entry timing (DTE target, IV condition to trigger entry)
  • Profit target: close at X% of max profit (e.g., 50% for iron condors, 75% for short spreads)
  • Stop-loss trigger
  • Adjustment plan if position moves against you

7. Upcoming Catalyst Calendar

  • Earnings date and expected move (if applicable)
  • Ex-dividend date (assignment risk for short calls)
  • Any scheduled corporate events (investor day, product launch, FDA decision)

Thesis Invalidation

After delivering the analysis signal, specify what would reverse it:

If signal is BULLISH — thesis breaks if:

  • IV Rank drops below 10 suggesting complacency AND put/call ratio inverts bearishly
  • [One or two more triggers drawn from this analysis's own drivers, each with a threshold]

If signal is BEARISH — thesis breaks if:

  • unusual call buying >3x average AND IV skew flips to call premium
  • [One or two more triggers drawn from this analysis's own drivers, each with a threshold]

Re-run this analysis when:

  • Next earnings release
  • Price moves ±15% from current level
  • 60 days have elapsed
  • Material news event (acquisition, leadership change, regulatory decision)

Standard Signal Output

All analysis concludes with this standardized block:

In the output, the box below is followed immediately by this machine-readable footer — the same values as the box, filled in (no placeholders, no comments), so tools can read the result without parsing the box:

{"investskill": "1", "skill": "options-analysis", "ticker": "<TICKER, or null for a portfolio / market-wide view>", "as_of": "<YYYY-MM-DD from Data & Sources>", "signal": "<BULLISH | NEUTRAL | BEARISH>", "confidence": "<HIGH | MEDIUM | LOW>", "horizon": "<SHORT-TERM | MEDIUM-TERM | LONG-TERM>", "score": 0.0, "action": "<BUY | HOLD | SELL>", "conviction": "<STRONG | MODERATE | WEAK>", "retrieval": "<pasted by user | web/tool retrieval | model memory>", "sources": ["<each primary source, dated>"]}
╔══════════════════════════════════════════════╗
║              INVESTMENT SIGNAL               ║
╠══════════════════════════════════════════════╣
║ Signal:      BULLISH / NEUTRAL / BEARISH     ║
║ Confidence:  HIGH / MEDIUM / LOW             ║
║ Horizon:     SHORT / MEDIUM / LONG-TERM      ║
║ Score:       X.X / 10                        ║
╠══════════════════════════════════════════════╣
║ Action:      BUY / HOLD / SELL               ║
║ Conviction:  STRONG / MODERATE / WEAK        ║
╚══════════════════════════════════════════════╝

Score Guide: 8.0–10.0 Strongly Bullish | 6.0–7.9 Moderately Bullish | 4.0–5.9 Neutral | 2.0–3.9 Moderately Bearish | 0.0–1.9 Strongly Bearish Confidence: HIGH (strong data, clear signals) | MEDIUM (mixed signals) | LOW (limited data, conflicting signals) Horizon: SHORT-TERM (1 week–3 months) | MEDIUM-TERM (3 months–1 year) | LONG-TERM (1+ years)

Disclaimer: Educational analysis only. Not financial advice.

Portfolio Review

📋 Data & Sources Header — Open Every Output With It

The first thing in the output is this provenance block, filled in — never left as placeholders. It is the standard documented on the Data & Accuracy page and the first thing result-validator looks for:

Data & Sources
  As of:      <date the figures represent, e.g. 2026-06-30>
  Source:     <primary docs — SEC EDGAR 10-K/10-Q, company IR, FRED, exchange data …>
  Retrieval:  <pasted by user | web/tool retrieval | model memory>
  Confidence: <HIGH | MEDIUM | LOW>
  • Retrieval: model memory must be paired with Confidence: LOW — memory is a placeholder until confirmed against a primary source.
  • Mixed sources: list each with its own as-of date rather than blending them.
  • Sources disagree on the same figure (a closing price, a deal term): prefer the primary source (SEC filing, company release, exchange data) over an aggregator, then the most recent dated one. Show both values with their sources and dates, say which you used and why, and never average them.
  • Data the user pasted is reported as pasted by user; do not upgrade its confidence beyond what the user's own source supports.

🌐 Output Language — --lang zh-TW

English is the default. With --lang zh-TW — or when the user writes in Traditional Chinese and asks for a Chinese answer — write the whole analysis in Traditional Chinese (繁體中文, Taiwan usage), keeping it machine-checkable:

  • Translate every heading, narrative paragraph, table header and bullet. Give each metric's English name in brackets the first time it appears — 自由現金流(Free Cash Flow, FCF) — so the reader can match it to the glossary and to English analyses.
  • Keep in English tickers, figures and units, SEC form and item names (10-K Item 7, Form 4), the four Data & Sources field labels (As of, Source, Retrieval, Confidence — the Retrieval value may be written in Chinese, e.g. 使用者提供), and the Investment Signal box and JSON footer exactly as specified — field names and the BULLISH / NEUTRAL / BEARISH, HIGH / MEDIUM / LOW, BUY / HOLD / SELL, STRONG / MODERATE / WEAK values. Tools parse them.
  • Title the invalidation section 論點失效條件(Thesis Invalidation), and close with the disclaimer in Chinese: 免責聲明: 僅供教育用途,非投資建議。
  • Use Taiwan financial terms: 殖利率、本益比、營業利益率、自由現金流、聯準會、選擇權、買權/賣權、除息日、預扣稅.

Comprehensive portfolio analysis and optimization recommendations.

Data Verification

Before finalizing the analysis, verify:

  • Position weights sum to 100% (or account for cash)
  • Return figures use total return (dividends reinvested)
  • Benchmark used matches portfolio's investment universe
  • Correlation estimates based on at least 1 year of data
  • Tax calculations use correct marginal rate for the investor
  • Factor scores sourced from consistent data provider
  • VaR calculation uses current portfolio weights, not stale data

Phase 1 — Portfolio Snapshot

Start with a high-level summary before diving into details:

  • Total portfolio value and cash position
  • Number of holdings and average position size
  • Date range of analysis
  • Investor profile: risk tolerance (conservative / moderate / aggressive), time horizon, income vs. growth objective

Phase 2 — Performance Analysis

  1. Absolute Returns

    • YTD, 1Y, 3Y, 5Y total return
    • Compare against relevant benchmarks (S&P 500, Russell 2000, custom blended benchmark)
    • Annualized return vs. benchmark — outperformance or underperformance in basis points
  2. Risk-Adjusted Returns

    • Sharpe ratio (excess return per unit of total risk)
    • Sortino ratio (penalizes only downside volatility)
    • Calmar ratio (annualized return / max drawdown)
    • Information ratio vs. benchmark
  3. Drawdown Analysis

    • Maximum drawdown: peak-to-trough magnitude and duration
    • Recovery period after each significant drawdown
    • Drawdown frequency over the analysis window
    • Compare portfolio drawdown vs. benchmark during market stress periods
  4. Volatility Metrics

    • Annualized standard deviation of returns
    • Beta vs. S&P 500 (market sensitivity)
    • Tracking error vs. benchmark
    • Upside capture / downside capture ratios

Phase 3 — Asset Allocation Review

  1. Current Allocation Breakdown

    • Asset class weights: equities, fixed income, cash, REITs, commodities, alternatives
    • Equity sub-breakdown: US large cap, US small/mid cap, international developed, emerging markets
    • Compare current weights to target/policy weights — flag deviations > 5%
  2. Geographic Diversification

    • US vs. international exposure (%)
    • Country concentration risk — any single country > 20%?
    • Currency exposure and hedging status
  3. Sector and Industry Weights

    • GICS sector breakdown vs. S&P 500 sector weights
    • Overweight / underweight sectors and the investment thesis behind each
    • Cyclical vs. defensive balance relative to current macro environment
  4. Market Cap Distribution

    • Large cap / mid cap / small cap split
    • Growth vs. value tilt (P/E, P/B relative to benchmark)
    • Quality factor exposure (ROE, balance sheet strength)

Phase 4 — Holdings Review

  1. Individual Position Analysis

    • Position size as % of portfolio — flag any > 10% single-stock concentration
    • Contribution to total return (winners vs. laggards)
    • Unrealized gain/loss and holding period (tax impact)
  2. Position Sizing Discipline

    • Largest 5 positions — are sizes justified by conviction and risk/reward?
    • Smallest positions — are they too small to be meaningful?
    • Kelly criterion or volatility-scaled sizing check
  3. Overlap and Redundancy

    • Identify pairs of holdings with > 0.7 correlation
    • ETF overlap analysis — holdings duplicated across ETFs
    • Factor overlap: are multiple positions expressing the same single bet?
  4. Cost Efficiency

    • Weighted average expense ratio for ETF/fund holdings
    • Estimated annual trading cost drag
    • Tax lot optimization — identify specific lots to sell for tax efficiency

Phase 5 — Concentration Risk Scoring

Score each concentration dimension independently, then aggregate:

Single-Stock Concentration

Position WeightScoreStatus
< 5%0OK
5%–10%1Monitor
10%–15%2Warning
15%–20%3High Risk
> 20%5Red Flag — immediate review required

Sector Concentration

Largest Sector WeightScoreStatus
< 20%0Diversified
20%–25%1Slight tilt
25%–30%2Elevated
30%–40%3Warning
> 40%5Concentration risk — rebalance urged

Factor Concentration

Factor Tilt ConditionScoreNote
Mixed factors0Balanced
Mild single-factor tilt1Acceptable with thesis
Strong single-factor tilt (>70% one factor)3Vulnerable to factor drawdown
Pure single-factor portfolio (>90%)5Maximum factor risk

Common factor concentration traps: all high-growth names (P/E > 40), all micro/small-cap, all rate-sensitive (long-duration bonds + REITs + utilities together).

Aggregate Concentration Score

  • 0–2: Well-diversified
  • 3–5: Moderate concentration — monitor
  • 6–9: High concentration — rebalance
  • 10+: Critical — urgent review

Phase 6 — Correlation & Diversification Analysis

Measuring Diversification Benefit

Portfolio volatility reduction vs. average asset volatility signals diversification quality. If portfolio vol ≈ average position vol, diversification benefit is near zero.

Diversification Ratio = (Weighted average individual vol) / (Portfolio vol)

  • Ratio > 1.5: Good diversification
  • Ratio 1.2–1.5: Moderate
  • Ratio < 1.2: Poor — holdings move together

Pairwise Correlation Heatmap Template

For each pair of top-10 holdings, estimate 1-year rolling correlation:

         AAPL   MSFT   GOOGL  AMZN   NVDA   JPM   XOM   GLD
AAPL     1.00   0.85   0.82   0.79   0.76   0.42  0.15  -0.10
MSFT     0.85   1.00   0.84   0.80   0.78   0.41  0.14  -0.12
GOOGL    0.82   0.84   1.00   0.81   0.73   0.39  0.12  -0.09
AMZN     0.79   0.80   0.81   1.00   0.70   0.37  0.10  -0.08
NVDA     0.76   0.78   0.73   0.70   1.00   0.35  0.11  -0.07
JPM      0.42   0.41   0.39   0.37   0.35   1.00  0.48   0.05
XOM      0.15   0.14   0.12   0.10   0.11   0.48  1.00   0.30
GLD     -0.10  -0.12  -0.09  -0.08  -0.07   0.05  0.30  1.00

Color code: > 0.75 = HIGH (red), 0.50–0.75 = MODERATE (yellow), < 0.50 = LOW (green)

Effective Number of Positions (ENP)

ENP = 1 / sum(wi²) where wi = weight of position i

Interpretation:

  • ENP ≈ actual N: equal-weight, low correlation — true diversification
  • ENP << actual N: dominated by a few large/correlated positions
  • Target ENP: at least 0.6 × actual number of holdings

Correlated Cluster Identification

Group holdings with pairwise correlation > 0.75 into clusters. Each cluster behaves as a single economic bet. Example clusters: Big Tech, Energy Majors, Regional Banks, Long-Duration Fixed Income.

Actionable rule: no single cluster should exceed 35% of portfolio weight.

Phase 7 — Tax-Loss Harvesting Opportunities

Screening view only. For lot-level mechanics — the 61-day wash-sale window, DRIP and IRA purchases, specific-ID vs. FIFO, short- vs. long-term treatment, and non-US investors — run tax-lens, and let it overrule this phase where they differ.

Identification Criteria

Screen all positions for:

  1. Unrealized loss > 10% of cost basis
  2. No purchase of the same or a substantially identical security in the 30 days before the planned sale, and none planned in the 30 days after (including DRIP reinvestments and purchases in IRAs) — that is what the wash-sale rule tests, not how long the position has been held
  3. Realized gains available to offset in the current tax year

Wash-Sale Rule (30-Day Rule)

Do not repurchase the same or substantially identical security within 30 days before or after the sale. Alternatives to maintain exposure during the 30-day window: sell XOM, buy CVX; sell QQQ, buy VGT; sell one S&P 500 ETF for another.

Net Tax Benefit Calculation

Gross Tax Saving = Realized Loss × Marginal Tax Rate
Transaction Cost  = 2 × (Commission + ½ × Bid-Ask Spread × Trade Size)   (sell the loser and buy the replacement)
Net Benefit       = Gross Tax Saving − Transaction Cost
Harvest if:       Net Benefit > 0 AND loss > $500 minimum threshold

Tax-Loss Harvest Decision Table

ConditionAction
Unrealized loss > 10%, gains to offset, loss > $500Harvest — sell and replace with correlated substitute
Unrealized loss 5–10%, gains to offsetEvaluate: transaction cost vs. tax saving
Unrealized loss > 10%, no offsetting gainsHold or carry forward — harvest at year-end if gains materialize
Bought more of it (or a substantially identical fund) within the last 30 days, or DRIP is onThe loss would be washed — wait until 31 days after the last purchase, or turn DRIP off first; run tax-lens for the lot-level timeline
Loss < $500Skip — transaction costs exceed benefit
Short-term gain offsetPriority harvest — short-term rates are higher

Annual Harvest Tracking

Maintain a log: position, cost basis, current value, loss amount, harvest date, replacement security, re-entry date. Review in October before year-end to maximize benefit.

Phase 8 — Factor Exposure Analysis

Six-Factor Portfolio Map

For each holding, estimate factor scores (1–5 scale, 3 = neutral):

FactorMeasurement ProxyLow (1–2)Neutral (3)High (4–5)
ValueP/E, P/B vs. sectorExpensiveFairCheap
GrowthRevenue/EPS CAGR 3Y< 5%5–15%> 15%
Momentum12-1 month returnBottom quartileMiddleTop quartile
QualityROE, Debt/EBITDALow ROE, high debtAverageHigh ROE, low debt
Low-Vol1Y realized volHigh volAverageLow vol
SizeMarket capMega-capMid-capSmall-cap

Portfolio-Level Factor Score

Weight each holding's factor score by portfolio weight. Compare to S&P 500 baseline (all factors = 3.0):

Factor      Portfolio Score   Benchmark   Active Tilt
Value           2.1              3.0       -0.9  (Growth tilt)
Growth          4.2              3.0       +1.2  (Strong growth)
Momentum        3.8              3.0       +0.8  (Slight momentum)
Quality         3.5              3.0       +0.5  (Slight quality)
Low-Vol         2.3              3.0       -0.7  (Higher volatility)
Size            2.4              3.0       -0.6  (Large-cap tilt)

Unintended Factor Tilt Detection

Flag when any active tilt exceeds ±1.0: this signals an unintended concentration that may not be in the investment thesis. Common unintended tilts: growth investors inadvertently loading up on momentum; dividend investors loading up on low-vol (rate sensitive); index investors with large-cap tilt through tech sector overweight.

Benchmark Factor Comparison

Use the factor tilts to explain return difference from benchmark. Growth tilt +1.2 in a value-outperforming market explains underperformance. Momentum tilt in a mean-reverting market increases drawdown risk.

Phase 9 — Rebalancing Decision Framework

When to Rebalance

Three trigger systems — use the one matching the investor's approach:

  1. Threshold-Based (Recommended for most investors)

    • Rebalance when any asset class drifts > 5% from target weight
    • Individual stock position drifts > 3% from target
    • No action if drift is within tolerance band
  2. Calendar-Based

    • Review quarterly; rebalance only if drift > 3%
    • Annual full rebalance regardless of drift
    • Simpler but may miss significant mid-quarter moves
  3. Factor-Based

    • Rebalance when portfolio factor score diverges > 1.5 from target
    • Triggered by momentum factor decay (top-quartile stocks reversing)
    • Suitable for factor-aware investors

Cost-Benefit Analysis

Before executing any rebalance, verify it clears the cost hurdle:

Expected Drift Cost (annual) = Excess vol from drift × Sharpe ratio penalty
Transaction Cost = Commission + ½ × Bid-Ask Spread × Trade Size + Tax Impact   (per trade; same spread convention as the harvest calculation)
Rebalance if: Drift Cost > Transaction Cost × 2 (safety margin)

Rule of thumb: do not rebalance positions with < $2,000 drift — transaction costs exceed benefit.

Rebalancing Decision Flowchart

START
  |
  v
Is drift > threshold?
  NO  --> Monitor next period
  YES --> Is account taxable?
            YES --> Are there harvesting opportunities?
                      YES --> Combine harvest + rebalance
                      NO  --> Evaluate tax cost of rebalance
                               Tax cost > drift benefit? --> Delay
                               Tax cost < drift benefit? --> Rebalance
            NO  --> Rebalance immediately (no tax friction)

Tax-Smart Rebalancing Order

  1. Use new cash contributions to buy underweight assets first
  2. Reinvest dividends into underweight positions
  3. Harvest losses in overweight positions to fund purchases
  4. If still needed, sell overweight positions (minimize short-term gains)

Phase 10 — Drawdown & Risk Budget Analysis

A first pass. For historical scenario replay (2008, March 2020, 2022, the 2025 tariff shock), parametric VaR/CVaR at 95/99 %, correlation-spike and rate/USD/oil sensitivity, and days-to-exit liquidity, run risk-stress-test.

Per-Position Maximum Drawdown

For each holding, calculate the worst peak-to-trough decline over the analysis window:

PositionMax DrawdownRecovery TimeCurrent From PeakRisk Level
Example-35%14 months-8%Medium

Flag positions with max drawdown > 50%: high-volatility securities require smaller position sizes to avoid outsized portfolio impact.

Portfolio-Level Value at Risk (VaR)

95% VaR (1-day): the loss not exceeded on 95% of trading days.

Parametric VaR (simplified) = Portfolio Value × Portfolio Daily Vol × 1.645

Example: $500,000 portfolio, daily vol 0.9% → 95% VaR = $500,000 × 0.009 × 1.645 = $7,403/day

Report also: 99% VaR (multiply by 2.326 / 1.645) and 10-day VaR (multiply by √10).

Contribution to Portfolio Volatility

Each position's marginal contribution to total portfolio volatility:

Marginal Vol Contribution (i) = weight(i) × Cov(i, portfolio) / Portfolio Vol
% Vol Contribution = Marginal Vol Contribution / Portfolio Vol × 100

Target: no single position contributes > 20% of total portfolio volatility unless it is a deliberate high-conviction overweight.

Risk Budget Allocation Table

PositionWeight %Vol Contribution %Risk Budget UsedStatus
Ideal≤ 10%≤ 15%≤ budgetOK
Flag> 10%> 20%Over budgetTrim

Stress Test Scenarios

Apply historical shock scenarios to estimate portfolio impact:

  • 2022 Rate Shock: tech -35%, bonds -15%, energy +50%
  • 2020 COVID Crash: market -34% in 33 days, recovery 5 months
  • 2008 GFC: market -57%, credit spreads +500bps
  • 2000 Dot-com: tech -78% over 2.5 years

For each scenario: estimate portfolio decline, identify worst-hit positions, confirm portfolio can withstand the scenario without forcing a distressed sale.

Phase 11 — Optimization Recommendations

  1. Rebalancing Actions

    • List all positions with > 2% drift from target weight
    • Priority order: largest drift first
    • Tax-aware rebalancing: prioritize tax-loss harvesting in taxable accounts
    • Proposed trades: ticker, action (buy/sell/trim), target weight, estimated proceeds
  2. Diversification Improvements

    • Specific gaps to fill (e.g., no international exposure, no inflation hedge)
    • Correlation-reduction swaps: replace redundant holdings with uncorrelated alternatives
    • Alternative asset candidates if appropriate (gold, TIPS, REITs, private credit proxies)
  3. Underperformer Review

    • For each position underperforming benchmark by > 10% over 1Y:
      • Is the original thesis intact?
      • Has fundamentals deteriorated or is this a temporary setback?
      • Hold / average down / exit decision with rationale
  4. Upgrade Candidates

    • Positions to add based on current conviction and valuation
    • New opportunities that fit the portfolio's existing factor tilts
    • Cost reduction opportunities (replace active fund with cheaper ETF equivalent)

Output Format

Deliver the review as a structured report with these sections:

  1. Executive Summary (3–5 bullet points: biggest strengths and risks)
  2. Performance Scorecard (table: metric | portfolio | benchmark | delta)
  3. Allocation Heat Map (sector/geography breakdown in table form)
  4. Concentration Risk Score (aggregate score with dimension breakdown)
  5. Factor Exposure Summary (six-factor table vs. benchmark)
  6. Holdings Table (position | weight | 1Y return | vol contribution % | status: hold/trim/exit)
  7. Tax-Loss Harvest Candidates (position | unrealized loss | estimated tax saving)
  8. Risk Budget Table (position | weight | vol contribution | status)
  9. Action List (prioritized, with expected portfolio impact)
  10. Implementation Roadmap (immediate / 30-day / 90-day actions)

Focus on actionable insights aligned with the investor's stated objectives and risk tolerance. Flag any positions or allocations that conflict with the investor profile.

Thesis Invalidation

A portfolio review is a plan, and plans expire. After the signal, state what would make this review's action list wrong:

If the signal is BULLISH (portfolio is well-positioned) — thesis breaks if:

  • Any single position breaches the concentration limit set in this review (state the % and the ticker)
  • Realized correlation between the top holdings rises above the level assumed in the risk-budget table (diversification that only exists on paper)
  • The investor's stated objective or horizon changes — a portfolio built for accumulation is wrong for drawdown

If the signal is BEARISH (portfolio needs repair) — thesis breaks if:

  • The flagged positions are trimmed or hedged as recommended and the risk-budget table re-balances
  • The concentration that drove the call resolves through price (the over-weight position falls) rather than through action — re-check before selling into weakness
  • The benchmark used turns out to be the wrong universe for the mandate

Re-run this review when:

  • Quarter-end (rebalance check)
  • Any position moves more than ±25% relative to the portfolio
  • A contribution or withdrawal larger than 10% of portfolio value
  • Tax-year end (harvest window)
  • The investor's objectives, income needs, or horizon change

Standard Signal Output

All analysis concludes with this standardized block:

In the output, the box below is followed immediately by this machine-readable footer — the same values as the box, filled in (no placeholders, no comments), so tools can read the result without parsing the box:

{"investskill": "1", "skill": "portfolio-review", "ticker": "<TICKER, or null for a portfolio / market-wide view>", "as_of": "<YYYY-MM-DD from Data & Sources>", "signal": "<BULLISH | NEUTRAL | BEARISH>", "confidence": "<HIGH | MEDIUM | LOW>", "horizon": "<SHORT-TERM | MEDIUM-TERM | LONG-TERM>", "score": 0.0, "action": "<BUY | HOLD | SELL>", "conviction": "<STRONG | MODERATE | WEAK>", "retrieval": "<pasted by user | web/tool retrieval | model memory>", "sources": ["<each primary source, dated>"]}
╔══════════════════════════════════════════════╗
║              INVESTMENT SIGNAL               ║
╠══════════════════════════════════════════════╣
║ Signal:      BULLISH / NEUTRAL / BEARISH     ║
║ Confidence:  HIGH / MEDIUM / LOW             ║
║ Horizon:     SHORT / MEDIUM / LONG-TERM      ║
║ Score:       X.X / 10                        ║
╠══════════════════════════════════════════════╣
║ Action:      BUY / HOLD / SELL               ║
║ Conviction:  STRONG / MODERATE / WEAK        ║
╚══════════════════════════════════════════════╝

Score Guide: 8.0–10.0 Strongly Bullish | 6.0–7.9 Moderately Bullish | 4.0–5.9 Neutral | 2.0–3.9 Moderately Bearish | 0.0–1.9 Strongly Bearish Confidence: HIGH (strong data, clear signals) | MEDIUM (mixed signals) | LOW (limited data, conflicting signals) Horizon: SHORT-TERM (1 week–3 months) | MEDIUM-TERM (3 months–1 year) | LONG-TERM (1+ years)

Disclaimer: Educational analysis only. Not financial advice.

Position Ladder — Staged Entry & Cost-Basis Management

⚠️ Data Verification — Do This Before Any Analysis

Before running any analysis, always retrieve the latest market data for the ticker:

  1. Fetch current price — use web search or ask the user for the live price, 52-week range, and market cap. Never assume a price from training data.
  2. Confirm key figures — the current price and recent ATR or volatility for rung spacing, the user's lots (dates, prices, shares), cash available, and the thesis invalidation level the ladder must respect.
  3. State your data source — fill in the Data & Sources header (next section) so the origin, as-of date, retrieval path, and confidence of every figure are explicit at the top of the output.
  4. Flag stale data explicitly — if live data is unavailable, display this warning before proceeding:

⚠️ Live data unavailable. The following analysis uses training-data estimates which may be significantly out of date. Verify all prices and metrics before making any decisions.

Never silently substitute training-data estimates for current prices. When in doubt, ask the user to paste the latest quote.


📋 Data & Sources Header — Open Every Output With It

The first thing in the output is this provenance block, filled in — never left as placeholders. It is the standard documented on the Data & Accuracy page and the first thing result-validator looks for:

Data & Sources
  As of:      <date the figures represent, e.g. 2026-06-30>
  Source:     <primary docs — SEC EDGAR 10-K/10-Q, company IR, FRED, exchange data …>
  Retrieval:  <pasted by user | web/tool retrieval | model memory>
  Confidence: <HIGH | MEDIUM | LOW>
  • Retrieval: model memory must be paired with Confidence: LOW — memory is a placeholder until confirmed against a primary source.
  • Mixed sources: list each with its own as-of date rather than blending them.
  • Sources disagree on the same figure (a closing price, a deal term): prefer the primary source (SEC filing, company release, exchange data) over an aggregator, then the most recent dated one. Show both values with their sources and dates, say which you used and why, and never average them.
  • Data the user pasted is reported as pasted by user; do not upgrade its confidence beyond what the user's own source supports.

🌐 Output Language — --lang zh-TW

English is the default. With --lang zh-TW — or when the user writes in Traditional Chinese and asks for a Chinese answer — write the whole analysis in Traditional Chinese (繁體中文, Taiwan usage), keeping it machine-checkable:

  • Translate every heading, narrative paragraph, table header and bullet. Give each metric's English name in brackets the first time it appears — 自由現金流(Free Cash Flow, FCF) — so the reader can match it to the glossary and to English analyses.
  • Keep in English tickers, figures and units, SEC form and item names (10-K Item 7, Form 4), the four Data & Sources field labels (As of, Source, Retrieval, Confidence — the Retrieval value may be written in Chinese, e.g. 使用者提供), and the Investment Signal box and JSON footer exactly as specified — field names and the BULLISH / NEUTRAL / BEARISH, HIGH / MEDIUM / LOW, BUY / HOLD / SELL, STRONG / MODERATE / WEAK values. Tools parse them.
  • Title the invalidation section 論點失效條件(Thesis Invalidation), and close with the disclaimer in Chinese: 免責聲明: 僅供教育用途,非投資建議。
  • Use Taiwan financial terms: 殖利率、本益比、營業利益率、自由現金流、聯準會、選擇權、買權/賣權、除息日、預扣稅.

⚠️ Read This First — This Is Scenario Modeling, Not a Recommendation

This framework is more prescriptive than the rest of the InvestSkill catalog: it outputs specific price levels and share counts. Every number it produces is a conditional plan under stated assumptions, not an instruction to trade.

Three honest claims must appear in every output:

  1. Lowering average cost is not the same as making money. Average cost is an accounting anchor and a psychological one — it is not a return metric. Trimming a high-cost lot mechanically lowers the number without creating a dollar of wealth. Always report total return and realized P&L alongside average cost, and show the buy-and-hold comparison.
  2. This is volatility harvesting. The strategy monetizes oscillation around a level. It outperforms in a range-bound / mean-reverting tape and underperforms in a strong trend — in either direction. A stock that only goes up leaves the ladder underfilled; a stock that only goes down fills it and keeps falling.
  3. Position control is the actual edge. The hard ceiling on share count — "stop adding at N shares, no matter what" — is what converts averaging down from an unbounded liability into a bounded, pre-sized bet. Without the cap, this is not a strategy; it is a mechanism for concentrating capital into losers.

State all three, then produce the plan. Never present the ladder as a way to guarantee a recovery.


Overview

Answer the question the rest of the catalog does not: "I own this — now what is my execution plan?"

The other frameworks decide what to own (stock-eval), what it is worth (stock-valuation), and when to enter (technical-analysis). None of them manages a position across its life. This one does. It takes a holding — or a planned holding — and produces:

  • A share-count band: a floor (core position you always keep) and a ceiling (the hard cap on adding).
  • An entry ladder: pre-committed rungs between today's price and the bottom of the build zone, with the size of each.
  • A trim/re-add cycle: sell the highest-cost lots when price trades above blended average cost, buy them back below it, oscillating inside the band.
  • The accounting truth: capital at full deployment, blended average cost at each stage, realized vs. unrealized P&L, wash-sale exposure, and the total-return comparison against simply holding.
  • A thesis-break gate: the conditions under which the correct action is to stop laddering and exit.

This is distinct from portfolio-review, which operates across holdings (allocation, drift, concentration, tax-loss harvesting). This one operates inside a single holding. Use portfolio-review to decide how big the position may be; use this skill to decide how to get there and how to manage it once there.


1. Inputs

Collect these before modeling. Ask for anything missing rather than assuming — the plan is only as good as the position cap.

InputRequiredDefault if unstated
Ticker✅—
Current shares held and average cost✅ (or 0 for a new position)—
Live market price✅fetch, or ask
Ceiling: max shares or max % of portfolio or max dollars✅ask — never invent a cap
Floor: the core position never sold—60% of ceiling
Portfolio size (for the concentration check)—ask; skip concentration scoring if withheld
Rung spacing method—1.0 × ATR(14), rounded to a clean price
Number of rungs—5
Account type: taxable / IRA / 401(k) / non-US—taxable (the conservative assumption — full tax friction)
Lot-accounting method: FIFO / specific-ID / average—FIFO (most brokers' default)
Asset type: single stock / broad ETF / leveraged ETF—single stock (full thesis gate applies)
Time budget: how long you will let the ladder work—2 quarters

2. Phase 1 — Position Control (控倉): Set the Bounds First

The bounds are set before the first share is bought, and they are the part of the plan that does not move.

Derive the ceiling — take the most restrictive of:

  • Concentration cap: single-position weight ≤ 10% of portfolio for a large-cap, ≤ 5% for a high-beta or single-product name (align with portfolio-review's flags).
  • Capital cap: total dollars you will commit if every rung fills. Compute this explicitly — the most common failure is discovering the full ladder costs more cash than exists.
  • Loss-tolerance cap: ceiling shares × (bottom rung − plausible bear-case price). If the position filled completely and then fell to the bear-case target, is the dollar loss survivable? If not, the ceiling is too high.

Derive the floor — the share count you would still want if the stock did nothing for two years. Typically 50–70% of ceiling. The floor exists so the trim leg can never sell you out of a name you want to own.

The cap rule (state it verbatim in the output):

At the ceiling, adding stops. Further weakness is not a reason to add beyond the cap — it is a reason to re-run the thesis gate in Phase 6. Raising the ceiling mid-drawdown is the single most common way this plan fails.


3. Phase 2 — Build the Entry Ladder

Rung spacing

MethodSpacingBest for
Volatility-scaled (default)1.0–1.5 × ATR(14) per rungAny name — spacing adapts to how much the stock actually moves
Fixed percentage3–4% (mega-cap), 5–7% (mid-cap), 8–12% (high-beta)Simple, mechanical, easy to place as resting orders
Support-basedRungs at identified support / value-area levelsWhen technical-analysis has produced a clean support table — the highest-quality placement
Fixed dollarEqual $ incrementsRound-number psychology; only sensible for a narrow price band

Rungs spaced tighter than ~0.5 × ATR will all fill on a single day's noise, which defeats the purpose. Rungs wider than ~2 × ATR rarely fill at all.

Rung sizing

SizingMechanicEffect on average cost
Equal sharesSame share count per rungSimplest; average cost = simple mean of rung prices
Equal dollars (preferred)Same dollar amount per rungMechanically buys more shares at lower prices → average cost is always ≤ equal-shares
PyramidSize increases at lower rungs (e.g. 1×, 1×, 1.5×, 2×, 2.5×)Lowest average cost, but concentrates the most capital into the deepest drawdown — requires the strongest thesis conviction
Inverted (anti-)pyramidSize decreases lowerDo not use for accumulation; it is a scaling-out pattern

Required ladder math

Always compute and show:

  • Blended average cost at full fill — Σ(rung price × rung shares) ÷ total shares.
  • Total capital at full fill — the cash the plan demands in the worst case.
  • Drawdown to full fill — (bottom rung ÷ current price) − 1. This is the decline the plan expects to sit through, stated up front.
  • Unrealized loss at full fill if price stops at the bottom rung — ceiling shares × (bottom rung − blended average). The number the user must be able to hold without abandoning the plan.
  • Dry powder remaining — capital not committed to this ladder, so the user sees the opportunity cost.

The underfill problem (and its fix)

A ladder on a name in a durable uptrend never fills. The position ends at 10–20% of target and the analysis was wasted — a real cost, not a hypothetical one. Two mitigations, both stated as explicit choices:

  • Starter tranche: deploy 30–40% of the target position at market immediately, ladder the remaining 60–70%. Guarantees meaningful exposure; costs you the better average if the dip arrives.
  • Time-based backstop: if the ladder is still under X% filled after the stated time budget and price is above the top rung, either re-anchor the ladder upward to current price or accept the smaller position. Decide which in advance — this is exactly the decision emotion makes badly in the moment.

4. Phase 3 — The Trim / Re-Add Cycle

Once the position is built, it oscillates inside the band.

Trim leg — when price trades above blended average cost:

  • Sell the highest-cost lots first, one lot per pre-set level, working down toward the floor.
  • Never sell below the floor share count.
  • Each trim level should sit at or above the cost basis of the lot being sold, so the trim is not a forced loss.
  • Recompute blended average cost after each trim and show it.

Re-add leg — when price trades below the new blended average cost:

  • Buy back toward the ceiling using the same rung discipline.
  • Re-add levels are set fresh from the current average cost, not the original ladder.
  • At the ceiling, stop. Again.

Cycle table — the primary deliverable of this phase:

StepTrigger priceActionShares ΔPositionBlended avg costCash ΔRealized P&L

Show at least one full cycle: build → trim to floor → re-add to ceiling. The end state of a completed cycle in a range-bound tape is the same share count at a lower average cost — that, and only that, is where the strategy's edge comes from.


5. Phase 4 — Lot Accounting, Taxes & Wash Sales

This is the phase most retail versions of this strategy omit, and it is where real money leaks. Treat it as mandatory for any taxable US account.

Lot selection matters

"Sell the highest-cost lot" is a specific-identification instruction. Under a broker's default FIFO, the oldest lot is sold instead — which may be the lowest-cost lot, producing the opposite tax outcome and a different average cost than the plan modeled. Flag this explicitly:

⚠️ This plan assumes specific-lot identification. Confirm your broker is not defaulting to FIFO, and select lots at the time of sale. FIFO will produce different realized P&L and a different post-trim average cost than modeled here.

Wash sales — the structural conflict in this strategy

The trim leg can realize a loss (selling a high-cost lot below its cost). The re-add leg then buys the same security back at a lower price. If the repurchase happens within 30 days before or after the loss sale, the loss is disallowed under the US wash-sale rule and is instead added to the basis of the replacement shares.

The trim/re-add cycle is, by construction, a wash-sale generator. Handle it:

  • Flag every trim that realizes a loss, and state the 30-day window in which a re-add would trigger a wash sale.
  • Note the practical consequence: the loss is deferred, not destroyed — basis moves to the new shares and the holding period carries over. Cash-flow timing suffers; the economics mostly do not.
  • Options: wait out the 31 days (risking the re-entry level), accept the deferral, or size the trim so it realizes a gain rather than a loss.
  • Wash-sale rules also reach across accounts, including a repurchase in an IRA — where the loss is permanently lost, not deferred. Warn on this specifically.

Holding period and account type

AccountTax frictionGuidance
TaxableFull: short/long-term capital gains, wash sales applyModel tax drag; prefer trims of lots held > 1 year; count the round-trip cost against the average-cost benefit
IRA / 401(k)None on internal tradesThe cleanest home for this strategy — cycle freely; note that realized losses are never deductible
Non-USVariesState that local rules were not modeled and must be checked

Every trim resets the holding-period clock on any re-added shares. A strategy that cycles quarterly will hold mostly short-term lots — a real, quantifiable cost in a taxable account that a naive average-cost calculation hides entirely.


6. Phase 5 — Reality Check: Total Return vs. Average Cost

Mandatory section. Produce this table for the completed plan under three price paths:

MetricBuy-and-hold (same capital)Ladder + trim/re-add
Shares held
Average cost
Realized P&L
Unrealized P&L
Total return ($ and %)
Est. tax drag

Run it for: (a) price recovers to the original entry, (b) price rallies well above the top rung, (c) price stays below the bottom rung.

The expected pattern — state it plainly:

  • Round trip (price returns to where it started): the cycler wins. Same shares, lower basis, positive return where buy-and-hold is flat.
  • Strong rally: the cycler loses, because the trim leg sold shares that then kept appreciating. Lower average cost, fewer shares, less money. Quantify the gap in dollars — do not describe it qualitatively.
  • Continued decline: both lose; the cycler loses more than a partial position but less than an uncapped averaging-down plan. The ceiling is what bounds the damage.

Also report opportunity cost: the capital parked waiting for lower rungs earned nothing (or T-bill yield) while committed to this plan.


7. Phase 6 — The Thesis-Break Gate

A ladder is only legitimate on a name still worth owning. Without this gate, disciplined averaging down is a machine for concentrating capital into deteriorating businesses.

Hard stop — abandon the ladder and consider exiting, do not add — if any of these fire:

  • The original investment thesis has been falsified, not merely delayed: structural margin compression, permanent demand loss, a broken product cycle, or the moat breached.
  • Accounting or governance red flags surface (run bear-case; a restatement, auditor change, or aggressive revenue recognition ends the ladder).
  • Leverage deterioration: covenant risk, a distressed refinancing, or dilutive emergency financing.
  • The decline is fundamental, not technical — consensus estimates are being cut as fast as the price is falling, so the stock is not getting cheaper.
  • The position has already breached the portfolio concentration cap.

Do-not-ladder list — asset types where this framework should refuse to produce a plan and say why:

  • Leveraged and inverse ETFs — daily rebalancing makes them path-dependent with structural decay; averaging down compounds the decay.
  • Binary-event names — single-drug biotechs, litigation-outcome stocks, pending all-cash acquisitions. There is no mean to revert to; the distribution is bimodal.
  • Anything where the bear case is solvency, not valuation.

Regime fit check — the strategy needs oscillation:

SignalFavorable (ladder works)Unfavorable (ladder misfires)
ADX(14)< 25 — no dominant trend> 30 — strong trend; ladder underfills or catches a falling knife
Price vs. MA200Oscillating around itExtended far above (underfill) or in confirmed breakdown below
ATR as % of price1.5–5% — enough movement to fill rungs< 1% — rungs never fill; > 8% — spacing must widen materially
Recent behaviorRange-bound, repeated tests of supportOne-directional, gap-driven

Ladder Suitability Score (0–10) — the skill's headline number:

ComponentPointsWhat earns full marks
Thesis integrity0–3Would you buy this fresh today, at this price, knowing nothing of your entry?
Regime fit0–2Range-bound, mean-reverting, oscillating around a level
Volatility adequacy0–2ATR wide enough that rungs realistically fill inside the time budget
Position headroom0–2Full ceiling still fits inside the portfolio concentration cap
Account / tax fit0–1Tax-advantaged account, or trim levels sit in gain territory
ScoreReading
8.0–10.0Ladder is appropriate — execute the plan as modeled
6.0–7.9Workable with a reduced ceiling and wider rungs
4.0–5.9Marginal — prefer a single sized entry over a ladder
0.0–3.9Do not ladder — the thesis, regime, or concentration test failed

8. Scenario Presets

Match the preset to the situation, state which was applied, and show the parameters used.

PresetFloor / CeilingRung spacingSizingThesis gateNotes
Core compounder60% / 100%3–4% or 1× ATREqual dollarsFull fundamental gateTrim only above average cost; long holds favor tax-efficient trims
Broad-index ETF autopilot50% / 100%4–5% or 1× ATREqual dollarsNone — no company thesis to breakCleanest use case; regime and cap checks still apply
High-beta single stock40% / 100%8–12% or 1.5× ATRPyramidStrict + hard stopSmaller ceiling %, wider rungs, mandatory bear-case
Starter + ladder (anti-underfill)40% at market, ladder the rest1× ATREqual dollarsFull gateFor names that rarely pull back
Underwater rescueCurrent holding / hard capSupport-basedEqual dollarsMandatory re-underwriteOnly proceed if the fresh-money test passes; otherwise this is an exit plan, not a ladder

9. Worked Example

Position: 20 shares bought at $128. Price now $122. Target band: floor 60 shares, ceiling 100. Equal shares, $3 rungs, taxable account, specific-lot ID.

Ladder

RungPriceSharesCostStatus
1$12820$2,560filled (the original entry)
2$12520$2,500fills now — above market
3$12220$2,440fills now — at market
4$11920$2,380pending
5$11620$2,320pending

Full fill: 100 shares, $12,200 capital, blended average $122.00. Drawdown to full fill is −4.9% from today's $122 (116 ÷ 122 − 1) — that is the further decline the plan is signing up to sit through, and $4,700 of the $12,200 (rungs 4–5) is still uncommitted. Equal-dollar sizing of the same $12,200 would instead yield ~100.1 shares at $121.85 — mechanically lower, always.

Trim to floor — price recovers: sell the $128 lot at $125, then the $125 lot at $128.

  • Cash back: (20 × $125) + (20 × $128) = $5,060
  • Realized: −$60 on the first lot, +$60 on the second = $0 net
  • Remaining: 60 shares at $122 / $119 / $116 → average $119.00
  • ⚠️ The $125 sale realized a $60 loss. Any repurchase within 30 days is a wash sale — the loss is disallowed and added to the replacement shares' basis.

Re-add to ceiling — price falls back; buy 20 at $117 and 20 at $115.

  • 100 shares, total cash out $11,780, average $117.80 (down from $122.00)

Reality check at three prices

Price pathBuy-and-hold 100 @ $122Ladder + cycle
Back to $122$0+$420 — same 100 shares, $4.20 lower basis
Rallies to $140 (trimmed, never re-added)+$1,800+$1,260 — $540 worse; 40 shares were sold into the rally
Stalls at $115−$700−$280 on 100 shares at $117.80

The middle row is the honest cost of the strategy: the lower average cost was purchased with forgone upside. The first row is the edge. Which row you get is decided by the tape, not by the discipline.


10. Input Formats

Format 1: Manage an existing position

User: /position-ladder AVGO — I hold 20 shares at $128, now $122. Target 60–100 shares.

Claude builds the ladder, the trim/re-add cycle, and the total-return comparison.

Format 2: Plan a new position from scratch

User: /position-ladder MSFT — new position, max 5% of a $200k portfolio, ETF-style autopilot.

Claude derives the ceiling from the concentration cap and lays out the rungs.

Format 3: Pasted holdings, multiple positions

User: /position-ladder — here are my holdings: [paste]. Which of these should I be laddering?

Claude scores each position's Ladder Suitability and produces a plan only for those that pass the gate.

11. Output Format

  1. Assumptions Stated — every input used, every default applied, and what was assumed because it was not supplied. First section, always.
  2. Position Snapshot — current shares, average cost, market price, unrealized P&L, weight in portfolio.
  3. Position Bounds — floor, ceiling, and which of the three caps was binding.
  4. Entry Ladder Table — rungs, sizes, capital at full fill, blended average, drawdown to full fill.
  5. Trim / Re-Add Cycle Table — one complete cycle with running average cost and realized P&L.
  6. Lot Accounting & Tax Notes — lot method assumed, wash-sale flags with dates, holding-period effects.
  7. Reality Check — total return vs. buy-and-hold across all three price paths, in dollars.
  8. Thesis-Break Gate — the gate checklist, regime table, and the conditions that end the plan.
  9. Ladder Suitability Score — 0–10 with the component breakdown.
  10. Investment Signal Block.

Round share counts to whole shares (or state that fractional shares are assumed) and round rung prices to levels that can actually be entered as resting orders.


Notes

  • Complements, does not replace: run stock-eval first to establish the thesis, technical-analysis for the support levels that make the best rungs, bear-case for the gate in Phase 6, and portfolio-review for the concentration cap that sets the ceiling.
  • The strategy is regime-dependent, and the regime is not knowable in advance. Present it as a bounded, rules-based way to build a position — not as an edge that works everywhere.
  • Do not raise the ceiling mid-drawdown. If the analysis concludes the position should be bigger, that is a new decision requiring a fresh thesis, not a ladder adjustment.
  • Fractional-share brokers change the math — equal-dollar sizing becomes exact. Say which assumption is in force.
  • Tax treatment is modeled at a high level for US taxable accounts only. It is not tax advice; recommend a tax professional for anything material.

Thesis Invalidation

After delivering the analysis signal, specify what would reverse it:

If signal is BULLISH — thesis breaks if:

  • Price closes below the bottom rung on above-average volume with fundamental deterioration
  • Consensus estimates are cut faster than the price falls (the stock is not getting cheaper)
  • The position reaches its ceiling and the thesis gate no longer passes

If signal is BEARISH — thesis breaks if:

  • The falsified thesis element is restored: margins re-expand, demand returns, guidance is raised
  • The regime turns range-bound again (ADX falls below 25) with fundamentals stabilized
  • Valuation resets low enough that a fresh-money buyer would want the full position today

Re-run this analysis when:

  • Next earnings release
  • Price moves ±15% from current level
  • Any rung fills or any trim executes
  • 60 days have elapsed
  • Material news event (acquisition, leadership change, regulatory decision)

Signal Output

This skill measures execution suitability, not price direction — so state the mapping explicitly. The Ladder Suitability Score drives the block: a high score means staged accumulation inside the band is appropriate and the plan should be executed as modeled; a low score means the thesis, regime, or concentration test failed and the correct action is to stop adding. Action: BUY here means "continue laddering within the stated cap" — never "buy without limit." Action: SELL means the gate failed, not that a short is warranted. Read the direction of the underlying stock from stock-eval or bear-case, not from this block.

All analysis concludes with this standardized block:

In the output, the box below is followed immediately by this machine-readable footer — the same values as the box, filled in (no placeholders, no comments), so tools can read the result without parsing the box:

{"investskill": "1", "skill": "position-ladder", "ticker": "<TICKER, or null for a portfolio / market-wide view>", "as_of": "<YYYY-MM-DD from Data & Sources>", "signal": "<BULLISH | NEUTRAL | BEARISH>", "confidence": "<HIGH | MEDIUM | LOW>", "horizon": "<SHORT-TERM | MEDIUM-TERM | LONG-TERM>", "score": 0.0, "action": "<BUY | HOLD | SELL>", "conviction": "<STRONG | MODERATE | WEAK>", "retrieval": "<pasted by user | web/tool retrieval | model memory>", "sources": ["<each primary source, dated>"]}
╔══════════════════════════════════════════════╗
║              INVESTMENT SIGNAL               ║
╠══════════════════════════════════════════════╣
║ Signal:      BULLISH / NEUTRAL / BEARISH     ║
║ Confidence:  HIGH / MEDIUM / LOW             ║
║ Horizon:     SHORT / MEDIUM / LONG-TERM      ║
║ Score:       X.X / 10                        ║
╠══════════════════════════════════════════════╣
║ Action:      BUY / HOLD / SELL               ║
║ Conviction:  STRONG / MODERATE / WEAK        ║
╚══════════════════════════════════════════════╝

Score Guide: 8.0–10.0 Strongly Bullish | 6.0–7.9 Moderately Bullish | 4.0–5.9 Neutral | 2.0–3.9 Moderately Bearish | 0.0–1.9 Strongly Bearish Confidence: HIGH (strong data, clear signals) | MEDIUM (mixed signals) | LOW (limited data, conflicting signals) Horizon: SHORT-TERM (1 week–3 months) | MEDIUM-TERM (3 months–1 year) | LONG-TERM (1+ years)

Note: The Score above is the Ladder Suitability Score, mapped onto the standard scale for cross-skill comparability. It rates how appropriate staged accumulation is right now — not how attractive the stock is. Pair it with /stock-eval for the directional view.

Disclaimer: Educational analysis only. Not financial advice. All price levels and share counts are scenario models under stated assumptions, not trade instructions.

Report Generator

Generate exportable HTML/PDF reports with interactive charts and visualizations from investment analysis data.

Report Templates

  1. Executive Summary Report

    • High-level overview (1-2 pages)
    • Key metrics and investment thesis
    • Critical charts only
    • Ideal for stakeholder presentations
  2. Comprehensive Analysis Report

    • Full detailed analysis (5-10 pages)
    • All data sections and visualizations
    • Methodology and assumptions
    • Complete investment case
  3. Portfolio Review Report

    • Multi-stock comparison
    • Portfolio allocation charts
    • Risk metrics and diversification analysis
    • Performance attribution

Visualization Components

Financial Charts

  • Line Charts: Revenue/earnings trends, price history
  • Bar Charts: Quarterly comparisons, segment breakdowns
  • Stacked Area: Balance sheet composition, revenue mix
  • Waterfall Charts: Cash flow analysis, earnings bridges
  • Pie Charts: Geographic/product segment revenue split

Valuation Charts

  • Comparison Charts: P/E, P/B, EV/EBITDA vs. peers and historical averages
  • DCF Sensitivity Tables: Value ranges under different scenarios
  • Multiple Expansion/Contraction: Historical valuation bands

Technical Charts

  • Price Charts: Candlesticks with moving averages overlay
  • Volume Bars: Trading volume patterns
  • Indicator Panels: RSI, MACD, Bollinger Bands
  • Support/Resistance: Key price levels annotated

HTML Report Structure

Generate standalone HTML file with the following structure:

<!DOCTYPE html>
<html lang="en">
<head>
    <meta charset="UTF-8">
    <meta name="viewport" content="width=device-width, initial-scale=1.0">
    <title>Investment Analysis Report - [TICKER] - [DATE]</title>

    <script src="https://cdn.jsdelivr.net/npm/chart.js@4.4.0/dist/chart.umd.min.js"></script>

    <style>
        @import url('https://fonts.googleapis.com/css2?family=Inter:wght@300;400;500;600;700;800&family=JetBrains+Mono:wght@400;600&display=swap');

        :root {
            --navy:          #0F172A;
            --navy-mid:      #1E293B;
            --sky:           #0284C7;
            --sky-light:     #E0F2FE;
            --teal:          #0D9488;
            --teal-light:    #CCFBF1;
            --emerald:       #059669;
            --emerald-light: #D1FAE5;
            --red:           #DC2626;
            --red-light:     #FEE2E2;
            --amber:         #D97706;
            --amber-light:   #FEF3C7;
            --slate:         #64748B;
            --slate-light:   #F1F5F9;
            --border:        #E2E8F0;
            --text-primary:  #0F172A;
            --text-secondary:#475569;
            --text-muted:    #94A3B8;
            --grad-hero:     linear-gradient(135deg, #0F172A 0%, #1E3A5F 55%, #0284C7 100%);
            --grad-accent:   linear-gradient(90deg, #0284C7, #0D9488);
            --shadow-sm:     0 1px 3px rgba(0,0,0,.08), 0 1px 2px rgba(0,0,0,.05);
            --shadow-md:     0 4px 16px rgba(0,0,0,.08), 0 2px 8px rgba(0,0,0,.04);
        }

        * { margin: 0; padding: 0; box-sizing: border-box; }

        body {
            font-family: 'Inter', -apple-system, BlinkMacSystemFont, sans-serif;
            line-height: 1.6;
            color: var(--text-primary);
            background: #F8FAFC;
        }

        /* ── HERO ── */
        .report-hero {
            background: var(--grad-hero);
            padding: 56px 48px 48px;
            position: relative;
            overflow: hidden;
        }
        .report-hero::before {
            content: '';
            position: absolute; top: -80px; right: -80px;
            width: 420px; height: 420px; border-radius: 50%;
            background: rgba(2,132,199,.12);
        }
        .report-hero::after {
            content: '';
            position: absolute; bottom: -60px; left: -40px;
            width: 300px; height: 300px; border-radius: 50%;
            background: rgba(13,148,136,.10);
        }
        .hero-inner { position: relative; z-index: 1; max-width: 1100px; margin: 0 auto; }
        .hero-badge {
            display: inline-flex; align-items: center; gap: 6px;
            background: rgba(2,132,199,.18);
            border: 1px solid rgba(2,132,199,.35);
            color: #7DD3FC;
            font-size: 11px; font-weight: 600; letter-spacing: 1.2px; text-transform: uppercase;
            padding: 4px 14px; border-radius: 20px; margin-bottom: 22px;
        }
        .hero-ticker {
            font-size: 72px; font-weight: 800; color: #fff;
            letter-spacing: -2px; line-height: 1;
            font-family: 'JetBrains Mono', monospace;
        }
        .hero-company { font-size: 19px; font-weight: 400; color: rgba(255,255,255,.65); margin-top: 8px; }
        .hero-rule {
            width: 48px; height: 3px;
            background: var(--grad-accent); border-radius: 2px;
            margin: 24px 0;
        }
        .hero-meta { display: flex; flex-wrap: wrap; gap: 28px; }
        .hero-meta-item { display: flex; flex-direction: column; gap: 3px; }
        .hero-meta-label {
            font-size: 9px; font-weight: 700; letter-spacing: 1.2px;
            text-transform: uppercase; color: rgba(255,255,255,.38);
        }
        .hero-meta-value { font-size: 13px; font-weight: 500; color: rgba(255,255,255,.82); }

        /* ── BODY ── */
        .report-body { max-width: 1100px; margin: 0 auto; padding: 40px 24px 80px; }

        /* ── TABLE OF CONTENTS ── */
        .toc {
            background: #fff; border: 1px solid var(--border);
            border-radius: 12px; padding: 28px 32px;
            margin-bottom: 32px; box-shadow: var(--shadow-sm);
        }
        .toc-label {
            font-size: 10px; font-weight: 700; letter-spacing: 1.6px;
            text-transform: uppercase; color: var(--text-muted); margin-bottom: 14px;
        }
        .toc-grid {
            list-style: none;
            display: grid; grid-template-columns: repeat(auto-fill, minmax(210px, 1fr)); gap: 8px 24px;
        }
        .toc-grid a {
            font-size: 13px; font-weight: 500; color: var(--sky);
            text-decoration: none; display: flex; align-items: center; gap: 6px;
        }
        .toc-grid a::before { content: '›'; font-size: 14px; color: var(--text-muted); }
        .toc-grid a:hover { color: var(--teal); }

        /* ── SECTION CARD ── */
        .section {
            background: #fff; border: 1px solid var(--border);
            border-radius: 12px; padding: 32px;
            margin-bottom: 20px; box-shadow: var(--shadow-sm);
            page-break-inside: avoid;
        }
        .section-head {
            display: flex; align-items: center; gap: 12px;
            margin-bottom: 24px; padding-bottom: 18px;
            border-bottom: 1px solid var(--border);
        }
        .section-icon {
            width: 36px; height: 36px; border-radius: 8px;
            background: var(--sky-light);
            display: flex; align-items: center; justify-content: center;
            font-size: 17px; flex-shrink: 0;
        }
        .section-title { font-size: 18px; font-weight: 700; color: var(--navy); letter-spacing: -.3px; }
        .section-sub   { font-size: 12px; color: var(--text-muted); margin-top: 2px; }
        .section-num {
            margin-left: auto;
            font-size: 11px; font-weight: 600;
            font-family: 'JetBrains Mono', monospace;
            color: var(--text-muted);
            background: var(--slate-light);
            padding: 2px 9px; border-radius: 4px;
        }

        /* ── METRIC CARDS ── */
        .metrics-grid {
            display: grid; grid-template-columns: repeat(auto-fit, minmax(175px, 1fr));
            gap: 16px; margin: 24px 0;
        }
        .metric-card {
            background: var(--slate-light); border: 1px solid var(--border);
            border-radius: 10px; padding: 20px;
            transition: box-shadow .2s;
        }
        .metric-card:hover { box-shadow: var(--shadow-md); }
        .metric-card.positive { border-top: 3px solid var(--emerald); }
        .metric-card.negative { border-top: 3px solid var(--red); }
        .metric-card.neutral  { border-top: 3px solid var(--sky); }
        .metric-card.warning  { border-top: 3px solid var(--amber); }
        .metric-label {
            font-size: 10px; font-weight: 600; letter-spacing: .8px;
            text-transform: uppercase; color: var(--text-muted); margin-bottom: 8px;
        }
        .metric-value {
            font-size: 26px; font-weight: 800; color: var(--navy);
            letter-spacing: -.5px;
            font-family: 'JetBrains Mono', monospace;
        }
        .metric-delta {
            font-size: 11px; font-weight: 600; margin-top: 6px;
            display: flex; align-items: center; gap: 3px;
        }
        .metric-delta.up   { color: var(--emerald); }
        .metric-delta.down { color: var(--red); }
        .metric-delta.flat { color: var(--text-muted); }
        .metric-ctx { font-size: 11px; color: var(--text-muted); margin-top: 4px; }

        /* ── CHART CONTAINER ── */
        .chart-wrap {
            background: #fff; border: 1px solid var(--border);
            border-radius: 10px; padding: 24px; margin: 20px 0;
            box-shadow: var(--shadow-sm);
        }
        .chart-label {
            font-size: 11px; font-weight: 600; letter-spacing: .6px;
            text-transform: uppercase; color: var(--text-muted); margin-bottom: 16px;
        }
        .chart-canvas { position: relative; height: 320px; }

        /* ── TABLES ── */
        .table-scroll { overflow-x: auto; border-radius: 8px; border: 1px solid var(--border); margin: 16px 0; }
        table { width: 100%; border-collapse: collapse; font-size: 13px; }
        thead th {
            background: var(--navy); color: rgba(255,255,255,.80);
            padding: 11px 14px; text-align: left;
            font-size: 10px; font-weight: 600; letter-spacing: .7px; text-transform: uppercase;
            white-space: nowrap;
        }
        tbody tr:nth-child(even) { background: var(--slate-light); }
        tbody tr:hover           { background: var(--sky-light); }
        tbody td { padding: 10px 14px; border-bottom: 1px solid var(--border); }
        tbody tr:last-child td   { border-bottom: none; }
        .td-pos  { color: var(--emerald); font-weight: 600; }
        .td-neg  { color: var(--red);     font-weight: 600; }
        .td-mono { font-family: 'JetBrains Mono', monospace; font-size: 12px; }

        /* ── CALLOUT / HIGHLIGHT ── */
        .callout {
            border-left: 4px solid var(--sky); background: var(--sky-light);
            border-radius: 0 8px 8px 0; padding: 14px 18px;
            margin: 14px 0; font-size: 14px; color: var(--text-secondary);
        }
        .callout.warn    { border-color: var(--amber); background: var(--amber-light); }
        .callout.danger  { border-color: var(--red);   background: var(--red-light);   }
        .callout.success { border-color: var(--emerald); background: var(--emerald-light); }

        /* ── TAGS ── */
        .tag { display: inline-block; padding: 2px 9px; border-radius: 4px; font-size: 11px; font-weight: 600; }
        .tag-buy  { background: var(--emerald-light); color: var(--emerald); }
        .tag-sell { background: var(--red-light);     color: var(--red);     }
        .tag-hold { background: var(--amber-light);   color: var(--amber);   }
        .tag-info { background: var(--sky-light);     color: var(--sky);     }

        /* ── INVESTMENT SIGNAL BOX ── */
        .signal-box {
            background: var(--navy); border-radius: 12px;
            padding: 32px 36px; margin: 32px 0;
            position: relative; overflow: hidden;
        }
        .signal-box::before {
            content: '';
            position: absolute; top: 0; left: 0; right: 0; height: 3px;
            background: var(--grad-accent);
        }
        .signal-box.bullish .signal-watermark { color: rgba( 52,211,153,.07); }
        .signal-box.bearish .signal-watermark { color: rgba(248,113,113,.07); }
        .signal-box.neutral .signal-watermark { color: rgba(148,163,184,.07); }
        .signal-watermark {
            position: absolute; right: 28px; top: 50%;
            transform: translateY(-50%);
            font-size: 108px; font-weight: 800; letter-spacing: -4px;
            font-family: 'JetBrains Mono', monospace;
            pointer-events: none; user-select: none;
        }
        .signal-eyebrow {
            font-size: 9px; font-weight: 700; letter-spacing: 2px;
            text-transform: uppercase; color: rgba(255,255,255,.35);
            margin-bottom: 10px;
        }
        .signal-verdict {
            font-size: 44px; font-weight: 800; letter-spacing: -1.5px;
            font-family: 'JetBrains Mono', monospace;
            margin-bottom: 24px;
        }
        .signal-verdict.bullish { color: #34D399; }
        .signal-verdict.bearish { color: #F87171; }
        .signal-verdict.neutral { color: #94A3B8; }
        .signal-stats {
            display: flex; flex-wrap: wrap; gap: 20px 36px;
        }
        .signal-stat { display: flex; flex-direction: column; gap: 4px; }
        .signal-stat-label {
            font-size: 9px; font-weight: 700; letter-spacing: 1.2px;
            text-transform: uppercase; color: rgba(255,255,255,.35);
        }
        .signal-stat-val { font-size: 14px; font-weight: 700; color: rgba(255,255,255,.88); }
        .score-track {
            background: rgba(255,255,255,.12);
            border-radius: 3px; height: 5px; margin-top: 7px; overflow: hidden;
        }
        .score-fill {
            height: 100%; border-radius: 3px;
            background: var(--grad-accent);
        }

        /* ── FOOTER ── */
        .report-footer {
            border-top: 1px solid var(--border);
            padding-top: 24px; margin-top: 48px;
            display: flex; gap: 20px; align-items: flex-start;
        }
        .footer-disc {
            font-size: 11px; color: var(--text-muted); line-height: 1.75; flex: 1;
        }
        .footer-brand {
            font-size: 11px; font-weight: 600; color: var(--text-muted);
            white-space: nowrap; text-align: right; line-height: 1.6;
        }

        /* ── PRINT ── */
        @media print {
            body { background: #fff; }
            .report-hero { -webkit-print-color-adjust: exact; print-color-adjust: exact; }
            .section { box-shadow: none; }
            .chart-canvas { height: 260px; }
        }

        /* ── RESPONSIVE ── */
        @media (max-width: 768px) {
            .report-hero { padding: 36px 24px 32px; }
            .hero-ticker { font-size: 48px; }
            .report-body { padding: 24px 16px 48px; }
            .section { padding: 20px; }
            .metrics-grid { grid-template-columns: repeat(2, 1fr); }
            .signal-watermark { display: none; }
        }
    </style>
</head>
<body>

    <!-- HERO HEADER -->
    <div class="report-hero">
        <div class="hero-inner">
            <div class="hero-badge">&#9679;&nbsp; Investment Analysis Report</div>
            <div class="hero-ticker">[TICKER]</div>
            <div class="hero-company">[Company Full Name] &nbsp;·&nbsp; [Exchange]</div>
            <div class="hero-rule"></div>
            <div class="hero-meta">
                <div class="hero-meta-item">
                    <span class="hero-meta-label">Report Date</span>
                    <span class="hero-meta-value">[DATE]</span>
                </div>
                <div class="hero-meta-item">
                    <span class="hero-meta-label">Report Type</span>
                    <span class="hero-meta-value">[TYPE]</span>
                </div>
                <div class="hero-meta-item">
                    <span class="hero-meta-label">Analyst</span>
                    <span class="hero-meta-value">InvestSkill Investment Analysis</span>
                </div>
                <div class="hero-meta-item">
                    <span class="hero-meta-label">Price at Report</span>
                    <span class="hero-meta-value">$[PRICE]</span>
                </div>
            </div>
        </div>
    </div>

    <div class="report-body">

        <!-- Table of Contents -->
        <div class="toc">
            <div class="toc-label">Contents</div>
            <ul class="toc-grid">
                <li><a href="#summary">Executive Summary</a></li>
                <li><a href="#financials">Financial Performance</a></li>
                <li><a href="#valuation">Valuation Analysis</a></li>
                <li><a href="#technical">Technical Analysis</a></li>
                <li><a href="#risks">Risk Factors</a></li>
                <li><a href="#signal">Investment Signal</a></li>
            </ul>
        </div>

        <!-- 01 Executive Summary -->
        <div class="section" id="summary">
            <div class="section-head">
                <div class="section-icon">📋</div>
                <div>
                    <div class="section-title">Executive Summary</div>
                    <div class="section-sub">Key investment thesis and highlights</div>
                </div>
                <span class="section-num">01</span>
            </div>

            [Summary content]

            <div class="metrics-grid">
                <div class="metric-card positive">
                    <div class="metric-label">Revenue (TTM)</div>
                    <div class="metric-value">$[X]B</div>
                    <div class="metric-delta up">▲ [X]% YoY</div>
                </div>
                <div class="metric-card positive">
                    <div class="metric-label">Net Income</div>
                    <div class="metric-value">$[X]B</div>
                    <div class="metric-delta up">▲ [X]% YoY</div>
                </div>
                <div class="metric-card neutral">
                    <div class="metric-label">P/E Ratio</div>
                    <div class="metric-value">[X]x</div>
                    <div class="metric-ctx">vs sector [X]x</div>
                </div>
                <div class="metric-card neutral">
                    <div class="metric-label">Market Cap</div>
                    <div class="metric-value">$[X]T</div>
                    <div class="metric-ctx">[Exchange]</div>
                </div>
            </div>
        </div>

        <!-- 02 Financial Performance -->
        <div class="section" id="financials">
            <div class="section-head">
                <div class="section-icon">📊</div>
                <div>
                    <div class="section-title">Financial Performance</div>
                    <div class="section-sub">Revenue, earnings, and margin trends</div>
                </div>
                <span class="section-num">02</span>
            </div>

            <div class="chart-wrap">
                <div class="chart-label">Revenue &amp; Net Income — Annual ($B)</div>
                <div class="chart-canvas">
                    <canvas id="revenueChart"></canvas>
                </div>
            </div>

            [Analysis content]
        </div>

        <!-- 03 Valuation -->
        <div class="section" id="valuation">
            <div class="section-head">
                <div class="section-icon">💰</div>
                <div>
                    <div class="section-title">Valuation Analysis</div>
                    <div class="section-sub">DCF, multiples, and comparable companies</div>
                </div>
                <span class="section-num">03</span>
            </div>
            [Valuation content]
        </div>

        <!-- 04 Technical Analysis -->
        <div class="section" id="technical">
            <div class="section-head">
                <div class="section-icon">📈</div>
                <div>
                    <div class="section-title">Technical Analysis</div>
                    <div class="section-sub">Price action, indicators, and key levels</div>
                </div>
                <span class="section-num">04</span>
            </div>
            [Technical content]
        </div>

        <!-- 05 Risk Factors -->
        <div class="section" id="risks">
            <div class="section-head">
                <div class="section-icon">⚠️</div>
                <div>
                    <div class="section-title">Risk Factors</div>
                    <div class="section-sub">Key risks and mitigating considerations</div>
                </div>
                <span class="section-num">05</span>
            </div>
            [Risk content]
        </div>

        <!-- Investment Signal -->
        <div class="signal-box bullish" id="signal">
            <div class="signal-watermark">BULLISH</div>
            <div class="signal-eyebrow">Investment Signal</div>
            <div class="signal-verdict bullish">BULLISH</div>
            <div class="signal-stats">
                <div class="signal-stat">
                    <span class="signal-stat-label">Confidence</span>
                    <span class="signal-stat-val">HIGH</span>
                </div>
                <div class="signal-stat">
                    <span class="signal-stat-label">Time Horizon</span>
                    <span class="signal-stat-val">LONG-TERM</span>
                </div>
                <div class="signal-stat">
                    <span class="signal-stat-label">Action</span>
                    <span class="signal-stat-val">BUY</span>
                </div>
                <div class="signal-stat">
                    <span class="signal-stat-label">Conviction</span>
                    <span class="signal-stat-val">STRONG</span>
                </div>
                <div class="signal-stat">
                    <span class="signal-stat-label">Score</span>
                    <span class="signal-stat-val">8.4 / 10</span>
                    <div class="score-track">
                        <div class="score-fill" style="width: 84%"></div>
                    </div>
                </div>
            </div>
        </div>

        <!-- Footer -->
        <div class="report-footer">
            <div class="footer-disc">
                <strong>Disclaimer:</strong> This report is generated by an AI assistant for informational purposes only
                and does not constitute financial advice, a solicitation, or an offer to buy or sell any security.
                Past performance is not indicative of future results. All investments carry risk.
                Consult a licensed financial advisor before making investment decisions.
            </div>
            <div class="footer-brand">
                InvestSkill<br>Investment Analysis
            </div>
        </div>

    </div><!-- /.report-body -->

    <script>
        // Revenue & Earnings combo chart
        const revenueCtx = document.getElementById('revenueChart').getContext('2d');
        new Chart(revenueCtx, {
            type: 'bar',
            data: {
                labels: ['2020', '2021', '2022', '2023', '2024'],
                datasets: [
                    {
                        label: 'Revenue ($B)',
                        data: [100, 120, 145, 175, 210],
                        backgroundColor: 'rgba(2,132,199,.75)',
                        borderRadius: 4,
                        order: 2
                    },
                    {
                        label: 'Net Income ($B)',
                        data: [22, 28, 35, 44, 56],
                        type: 'line',
                        borderColor: '#0D9488',
                        backgroundColor: 'rgba(13,148,136,.10)',
                        borderWidth: 2.5,
                        pointRadius: 4,
                        pointBackgroundColor: '#0D9488',
                        tension: 0.35,
                        fill: true,
                        order: 1,
                        yAxisID: 'y1'
                    }
                ]
            },
            options: {
                responsive: true,
                maintainAspectRatio: false,
                interaction: { mode: 'index', intersect: false },
                plugins: {
                    legend: {
                        position: 'top',
                        labels: { font: { family: 'Inter', size: 11 }, usePointStyle: true, boxWidth: 8 }
                    },
                    tooltip: {
                        backgroundColor: '#0F172A',
                        titleColor: '#94A3B8',
                        bodyColor: '#FFFFFF',
                        borderColor: '#334155',
                        borderWidth: 1,
                        padding: 10,
                        cornerRadius: 6
                    }
                },
                scales: {
                    x: {
                        grid: { display: false },
                        ticks: { font: { family: 'Inter', size: 11 }, color: '#94A3B8' }
                    },
                    y: {
                        grid: { color: '#F1F5F9' },
                        ticks: { font: { family: 'Inter', size: 11 }, color: '#94A3B8' },
                        title: { display: true, text: 'Revenue ($B)', color: '#94A3B8', font: { size: 10 } }
                    },
                    y1: {
                        position: 'right',
                        grid: { display: false },
                        ticks: { font: { family: 'Inter', size: 11 }, color: '#0D9488' },
                        title: { display: true, text: 'Net Income ($B)', color: '#0D9488', font: { size: 10 } }
                    }
                }
            }
        });
    </script>
</body>
</html>

Chart Data Format

When receiving analysis data, structure chart data as follows:

// Line chart data structure
{
    chartType: 'line',
    title: 'Revenue Growth Trend',
    xAxis: ['2020', '2021', '2022', '2023', '2024'],
    datasets: [
        {
            label: 'Revenue ($B)',
            data: [100, 120, 145, 175, 210],
            color: '#0284C7'   // sky — primary series
        }
    ]
}

// Bar chart with multiple series
{
    chartType: 'bar',
    title: 'Quarterly Earnings Comparison',
    xAxis: ['Q1', 'Q2', 'Q3', 'Q4'],
    datasets: [
        {
            label: '2023',
            data: [1.2, 1.5, 1.8, 2.1],
            color: '#0284C7'   // sky — primary
        },
        {
            label: '2024',
            data: [1.5, 1.8, 2.2, 2.5],
            color: '#0D9488'   // teal — secondary
        }
    ]
}

PDF Export Instructions

To export the generated HTML report to PDF:

Method 1: Browser Print-to-PDF

  1. Open the HTML file in a web browser
  2. Press Ctrl+P (Cmd+P on Mac) or File → Print
  3. Select "Save as PDF" or "Microsoft Print to PDF"
  4. Adjust settings: margins, orientation, scale
  5. Save the PDF file

Method 2: wkhtmltopdf (Command Line)

# Install wkhtmltopdf
# macOS: brew install wkhtmltopdf
# Linux: apt-get install wkhtmltopdf

# Convert HTML to PDF
wkhtmltopdf \
  --page-size A4 \
  --margin-top 15mm \
  --margin-bottom 15mm \
  --enable-local-file-access \
  report.html report.pdf

Method 3: Playwright/Puppeteer (Node.js)

const { chromium } = require('playwright');

(async () => {
  const browser = await chromium.launch();
  const page = await browser.newPage();
  await page.goto('file:///path/to/report.html');
  await page.pdf({
    path: 'report.pdf',
    format: 'A4',
    printBackground: true
  });
  await browser.close();
})();

Integration with Other Skills

Input Formats Accepted

  1. Structured JSON Data

    {
      "ticker": "AAPL",
      "companyName": "Apple Inc.",
      "analysisType": "fundamental",
      "metrics": {...},
      "charts": [...]
    }
    
  2. Markdown Analysis Output

    • Parse markdown from stock-eval, technical-analysis, or portfolio-review
    • Extract tables, metrics, and key findings
    • Generate appropriate visualizations
  3. Custom Data Tables

    • CSV-formatted financial data
    • Excel/spreadsheet data ranges
    • Database query results

Workflow Examples

Example 1: Fundamental Analysis Report

User: /stock-eval AAPL
[Assistant generates analysis]

User: /report-generator --type comprehensive --format html
[Claude generates HTML report with embedded fundamental analysis and charts]

Example 2: Portfolio Review with Custom Data

User: /report-generator --type portfolio --data [paste portfolio data]
[Claude generates multi-stock comparison report with allocation charts]

Output

Generate a complete, standalone HTML file containing:

  1. File Metadata

    • Filename format: investment-report-[TICKER]-[YYYYMMDD]-[HHMMSS].html
    • File size indication
    • Estimated PDF size after export
  2. Report Contents

    • Professional header with company name, ticker, date
    • Table of contents (for comprehensive reports)
    • All analysis sections with appropriate formatting
    • Interactive charts with Chart.js
    • Data tables with styling
    • Key metrics highlighted in cards
    • Footer with disclaimer and generation info
  3. Technical Details

    • Inline CSS (no external dependencies except CDN for Chart.js)
    • Responsive design (mobile and desktop)
    • Print-optimized styling for PDF export
    • Cross-browser compatible (Chrome, Firefox, Safari, Edge)
  4. Export Instructions

    • Step-by-step guide for converting to PDF
    • Command-line examples for automated workflows
    • Browser settings recommendations

Usage Notes

  • The HTML file is fully self-contained except for Chart.js CDN (can be inlined if needed)
  • Charts are interactive in HTML (hover for data points, zoom, pan)
  • PDF export flattens charts to static images
  • For large portfolios (>20 stocks), consider pagination or summary views
  • Color scheme follows professional financial report standards (blue for primary, green for positive, red for negative)
  • All monetary values should be formatted with proper thousands separators and currency symbols

Integration with Signal Format

When generating reports from other skills, the report MUST prominently display each skill's Investment Signal block in a styled box. Use the following HTML (matches the CSS in the template above):

<!-- Bullish signal example — swap class to "bearish" or "neutral" as needed -->
<div class="signal-box bullish">
  <div class="signal-watermark">BULLISH</div>
  <div class="signal-eyebrow">Investment Signal</div>
  <div class="signal-verdict bullish">BULLISH</div>
  <div class="signal-stats">
    <div class="signal-stat">
      <span class="signal-stat-label">Confidence</span>
      <span class="signal-stat-val">HIGH</span>
    </div>
    <div class="signal-stat">
      <span class="signal-stat-label">Time Horizon</span>
      <span class="signal-stat-val">LONG-TERM</span>
    </div>
    <div class="signal-stat">
      <span class="signal-stat-label">Action</span>
      <span class="signal-stat-val">BUY</span>
    </div>
    <div class="signal-stat">
      <span class="signal-stat-label">Conviction</span>
      <span class="signal-stat-val">STRONG</span>
    </div>
    <div class="signal-stat">
      <span class="signal-stat-label">Score</span>
      <span class="signal-stat-val">X.X / 10</span>
      <div class="score-track">
        <div class="score-fill" style="width: XX%"></div>
      </div>
    </div>
  </div>
</div>

Design notes:

  • .signal-box class: deep navy background with teal/sky gradient top border
  • Add bullish, bearish, or neutral modifier class to both .signal-box and .signal-verdict to apply the correct verdict color and watermark tint
  • Score progress bar uses the score-fill width as a percentage (e.g. style="width: 84%" for 8.4/10)
  • Large ghost watermark text (.signal-watermark) reinforces the verdict at a glance

For multi-skill bundle reports (like /full-report), show a composite signal summary at the top of the report with individual skill scores in a row of .signal-stat items inside a single .signal-box.


When this skill generates or embeds a signal in an HTML report, the signal block MUST follow this standardized format:

╔══════════════════════════════════════════════╗
║              INVESTMENT SIGNAL               ║
╠══════════════════════════════════════════════╣
║ Signal:      BULLISH / NEUTRAL / BEARISH     ║
║ Confidence:  HIGH / MEDIUM / LOW             ║
║ Horizon:     SHORT / MEDIUM / LONG-TERM      ║
║ Score:       X.X / 10                        ║
╠══════════════════════════════════════════════╣
║ Action:      BUY / HOLD / SELL               ║
║ Conviction:  STRONG / MODERATE / WEAK        ║
╚══════════════════════════════════════════════╝

Score Guide: 8.0–10.0 Strongly Bullish | 6.0–7.9 Moderately Bullish | 4.0–5.9 Neutral | 2.0–3.9 Moderately Bearish | 0.0–1.9 Strongly Bearish Confidence: HIGH (strong data, clear signals) | MEDIUM (mixed signals) | LOW (limited data, conflicting signals) Horizon: SHORT-TERM (1 week–3 months) | MEDIUM-TERM (3 months–1 year) | LONG-TERM (1+ years)

Disclaimer: Educational analysis only. Not financial advice.

Research Bundle

This skill has been unified into full-report. Use /us-stock-analysis:full-report with a --depth flag instead:

Old UsageNew Equivalent
/research-bundle AAPL --quick/us-stock-analysis:full-report AAPL --depth quick
/research-bundle AAPL/us-stock-analysis:full-report AAPL --depth standard
/research-bundle AAPL (full)/us-stock-analysis:full-report AAPL --depth comprehensive

The full-report skill absorbs everything research-bundle provided: the 5-phase framework, composite scoring weights, investment thesis narrative, conflict resolution rules, entry/exit strategy, and monitoring plan — plus HTML report output and configurable depth.


Composite Scoring Weights (Quick Reference)

Component                Weight    Sub-Score (0-10)
─────────────────────────────────────────────────────
Business Quality          25%      [competitor + fundamental]
Valuation                 25%      [DCF + relative valuation]
Market Signals            20%      [insider + institutional + earnings]
Technical Setup           15%      [technical analysis timing]
Risk Profile              15%      [short interest + options signals]
─────────────────────────────────────────────────────
COMPOSITE SCORE          100%      X.X / 10

8.0–10.0  → Strong Buy   |  6.5–7.9  → Buy
5.0–6.4   → Hold/Watch   |  3.5–4.9  → Underweight
0.0–3.4   → Sell/Avoid

5-Phase Research Framework (reference)

The original research-bundle framework — now executed automatically by full-report:

Phase 1 — Business Foundation (competitor-analysis, sector-analysis) Assess economic moat width (Wide/Narrow/None), Porter's Five Forces, market share trajectory, and sector rotation positioning. Score: business quality 0–10.

Phase 2 — Valuation (dcf-valuation, stock-valuation) DCF intrinsic value across 3 scenarios + comparable company multiples (EV/EBITDA, P/E, P/S). Probability-weighted composite IV. Upside/downside to fair value.

Phase 3 — Market Signals (insider-trading, institutional-ownership, earnings-call-analysis) Net insider sentiment (Form 4 last 90 days), institutional accumulation/distribution (13F delta), management tone and guidance delta from last earnings call.

Phase 4 — Technical Setup (technical-analysis) MA stack (30/60/90/200/365d), RSI, MACD, Ichimoku, multi-timeframe confluence. Entry price, target, stop-loss, R:R ratio. Timing signal for position entry.

Phase 5 — Risk & Options (short-interest, options-analysis, economics-analysis) Short squeeze score, IV rank, put/call ratio skew, macro regime (yield curve, recession probability). Risk-adjusted position sizing.

Conflict Resolution Rules

When skills disagree, apply these priority overrides:

  • Technical SELL + Fundamental BUY → HOLD (wait for technical confirmation)
  • Insider BUY + Short Interest extreme (>20%) → CAUTION (investigate divergence)
  • DCF undervalued 30%+ + Technical BULLISH → STRONG BUY (conviction upgrade)
  • Macro recession >60% → override all bullish signals → REDUCE/AVOID
  • 3+ skills unanimous → allow confidence upgrade (MEDIUM → HIGH)

Depth Modes in full-report

  • --depth quick (5 modules): stock-eval, technical-analysis, dcf-valuation, insider-trading, earnings-call-analysis
  • --depth standard (10 modules): adds institutional-ownership, competitor-analysis, sector-analysis, options-analysis, short-interest
  • --depth comprehensive (15 modules): adds fundamental-analysis, stock-valuation, economics-analysis, financial-report-analyst, dividend-analysis

In the output, the box below is followed immediately by this machine-readable footer — the same values as the box, filled in (no placeholders, no comments), so tools can read the result without parsing the box:

{"investskill": "1", "skill": "research-bundle", "ticker": "<TICKER, or null for a portfolio / market-wide view>", "as_of": "<YYYY-MM-DD from Data & Sources>", "signal": "<BULLISH | NEUTRAL | BEARISH>", "confidence": "<HIGH | MEDIUM | LOW>", "horizon": "<SHORT-TERM | MEDIUM-TERM | LONG-TERM>", "score": 0.0, "action": "<BUY | HOLD | SELL>", "conviction": "<STRONG | MODERATE | WEAK>", "retrieval": "<pasted by user | web/tool retrieval | model memory>", "sources": ["<each primary source, dated>"]}
╔══════════════════════════════════════════════╗
║              INVESTMENT SIGNAL               ║
╠══════════════════════════════════════════════╣
║ Signal:      BULLISH / NEUTRAL / BEARISH     ║
║ Confidence:  HIGH / MEDIUM / LOW             ║
║ Horizon:     SHORT / MEDIUM / LONG-TERM      ║
║ Score:       X.X / 10                        ║
╠══════════════════════════════════════════════╣
║ Action:      BUY / HOLD / SELL               ║
║ Conviction:  STRONG / MODERATE / WEAK        ║
╚══════════════════════════════════════════════╝

Score Guide: 8.0–10.0 Strongly Bullish | 6.0–7.9 Moderately Bullish | 4.0–5.9 Neutral | 2.0–3.9 Moderately Bearish | 0.0–1.9 Strongly Bearish Confidence: HIGH (strong data, clear signals) | MEDIUM (mixed signals) | LOW (limited data, conflicting signals) Horizon: SHORT-TERM (1 week–3 months) | MEDIUM-TERM (3 months–1 year) | LONG-TERM (1+ years)

Disclaimer: Educational analysis only. Not financial advice.

Investment Result Validator

You are a rigorous meta-analyst. Your job is to critically evaluate the output of any InvestSkill analysis and produce a structured confidence assessment that tells the user how much to trust the conclusions.

How to Use

Paste or reference any prior analysis output (from /stock-eval, /stock-valuation, /technical-analysis, /bear-case, or any other skill). The validator will audit it across five dimensions and produce a Confidence Score Report.

Check the contract first. Before scoring, confirm the analysis carries the three pieces every InvestSkill skill is required to emit: a Data & Sources header at the top (As of · Source · Retrieval · Confidence), a Thesis Invalidation section, and the standard Investment Signal block. A missing Data & Sources header caps Data Quality at 10/20 (the sources cannot be verified); a Retrieval: model memory line with a Confidence above LOW is an automatic red flag; a missing Thesis Invalidation section scores 0 of 7 for "Bear case scenario modeled" in Dimension 4 (no stated way for the call to be wrong).


🌐 Output Language — --lang zh-TW

English is the default. With --lang zh-TW — or when the user writes in Traditional Chinese and asks for a Chinese answer — write the whole analysis in Traditional Chinese (繁體中文, Taiwan usage), keeping it machine-checkable:

  • Translate every heading, narrative paragraph, table header and bullet. Give each metric's English name in brackets the first time it appears — 自由現金流(Free Cash Flow, FCF) — so the reader can match it to the glossary and to English analyses.
  • Keep in English tickers, figures and units, SEC form and item names (10-K Item 7, Form 4), the four Data & Sources field labels (As of, Source, Retrieval, Confidence — the Retrieval value may be written in Chinese, e.g. 使用者提供), and the Investment Signal box and JSON footer exactly as specified — field names and the BULLISH / NEUTRAL / BEARISH, HIGH / MEDIUM / LOW, BUY / HOLD / SELL, STRONG / MODERATE / WEAK values. Tools parse them.
  • Title the invalidation section 論點失效條件(Thesis Invalidation), and close with the disclaimer in Chinese: 免責聲明: 僅供教育用途,非投資建議。
  • Use Taiwan financial terms: 殖利率、本益比、營業利益率、自由現金流、聯準會、選擇權、買權/賣權、除息日、預扣稅.

Validation Framework

Dimension 1: Data Quality (0–20 pts)

Evaluate the underlying data used in the analysis:

CheckPointsNotes
Data sources cited or identifiable0–5Named sources score higher. If a fact-check ledger is available, the whole Data Quality dimension (this table, out of 20) may not exceed what its Verification Score supports (≥ 8.0 → up to 20; 6.0–7.9 → up to 15; < 6.0 → up to 10)
Data recency (how fresh?)0–5<30 days = 5, 30–90 days = 3, >90 days = 1
Data completeness (missing fields?)0–5Count unfilled table cells, blanks, "N/A"
Data consistency (no contradictions)0–5Flag any internal conflicts

Data Quality Score: __ / 20


Dimension 2: Methodology Soundness (0–20 pts)

CheckPointsNotes
Appropriate valuation method for sector/stage0–5DCF for stable; P/S for growth; P/B for banks
Assumptions stated explicitly0–5Growth rate, WACC, terminal value, etc.
Assumptions within reasonable range0–5Compare to analyst consensus or historical norms
Multiple methods used / cross-validated0–52+ methods = full points

Methodology Score: __ / 20


Dimension 3: Signal Consistency (0–20 pts)

CheckPointsNotes
Technical and fundamental signals aligned0–7Same direction = 7, mixed = 3, opposing = 0
Sentiment signals (insider, institutional) aligned0–7Same direction = 7, mixed = 3, opposing = 0
Macro/sector context addressed and consistent with the thesis0–6Context assessed and it supports the thesis direction (tailwinds for a bull case, headwinds for a bear case) = 6; mixed or only partly addressed = 3; contradicts the thesis without acknowledgment, or ignored = 0. Never reward a bullish thesis for macro tailwinds a bearish one would be penalised for

Signal Consistency Score: __ / 20


Dimension 4: Risk Coverage (0–20 pts)

CheckPointsNotes
Key downside risks identified0–7At least 3 specific risks named
Bear case scenario modeled0–7Quantified, not just listed
Catalysts (positive + negative) identified0–6Upcoming events that could move the stock

Risk Coverage Score: __ / 20


Dimension 5: Reasoning Transparency (0–20 pts)

CheckPointsNotes
Conclusion follows logically from evidence0–7No unsupported leaps
Contrarian arguments considered0–7"Bull case even if wrong because…"
Limitations of the analysis acknowledged0–6Honest about what's unknown

Reasoning Transparency Score: __ / 20


Composite Confidence Score

Data Quality:            __ / 20
Methodology Soundness:   __ / 20
Signal Consistency:      __ / 20
Risk Coverage:           __ / 20
Reasoning Transparency:  __ / 20
─────────────────────────────────
TOTAL CONFIDENCE:        __ / 100

Confidence Tier:

  • 85–100: VERY HIGH — Analysis is thorough, data is fresh, signals align. Act with confidence.
  • 70–84: HIGH — Solid analysis with minor gaps. Good basis for a decision.
  • 55–69: MEDIUM — Usable but has notable gaps. Supplement with additional research.
  • 40–54: LOW — Significant data or methodology weaknesses. Treat conclusions as directional only.
  • 0–39: VERY LOW — Major issues found. Do not rely on this analysis without substantial rework.

Flags & Red Flags

List specific issues found during validation. Always check the output contract first — these are mechanical and come before any judgement:

Contract checkIf it fails
Data & Sources header present with As of · Source · Retrieval · Confidence🚩 Red flag; Data Quality capped at 10 / 20
Retrieval: model memory paired with anything but Confidence: LOW🚩 Red flag
Signal box complete (six fields) and its Score maps to its Signal by the Score Guide (skill-specific rules such as fact-check's Verification Score excepted)🚩 Red flag
JSON footer (```json {"investskill": …} right after the box) present⚠ Warning if missing — outputs from before the footer existed are still valid
JSON footer values (signal, score, confidence, action, conviction) equal the box🚩 Red flag if they disagree — one of the two was edited after the other

⚠ Warnings (moderate concern)

  • [List each warning]

🚩 Red Flags (serious concern)

  • [List each red flag]

✅ Strengths

  • [List what the analysis does well]

Recommended Next Steps

Based on the confidence score and flags, suggest:

  1. Which dimension(s) to strengthen first
  2. Which additional skills to run (e.g., "Run /stock-valuation to cross-check the valuation multiple")
  3. Specific data points to verify or refresh — for claim-level verification with citations, run fact-check on the report and feed its ledger back here

Adjusted Signal (if original had one)

If the validated analysis contained an Investment Signal block, reproduce it below with a confidence adjustment applied:

╔══════════════════════════════════════════════╗
║         VALIDATED INVESTMENT SIGNAL          ║
╠══════════════════════════════════════════════╣
║ Original Signal:  BULLISH / NEUTRAL / BEARISH║
║ Original Score:   X.X / 10                   ║
╠══════════════════════════════════════════════╣
║ Confidence Tier:  VERY HIGH / HIGH / MEDIUM  ║
║                   / LOW / VERY LOW           ║
║ Confidence Score: XX / 100                   ║
╠══════════════════════════════════════════════╣
║ Adjusted Action:  BUY / HOLD / SELL          ║
║ (Signal stands / weakened / reversed by      ║
║  low confidence — see flags above)           ║
╚══════════════════════════════════════════════╝

Output Format

Always end with the standard signal block reflecting the validation result itself (not the original analysis). Fill it from the validation, not by judgement:

Box fieldHow it is set
Score5.0 + (Original Score − 5.0) × Confidence Score / 100 — a low-confidence analysis is pulled toward neutral. No original signal → 5.0
SignalFrom that Score by the Score Guide (≥ 6.0 BULLISH · 4.0–5.9 NEUTRAL · < 4.0 BEARISH)
ConfidenceConfidence Tier collapsed to three: VERY HIGH / HIGH → HIGH · MEDIUM → MEDIUM · LOW / VERY LOW → LOW
HorizonThe original analysis' horizon
ActionThe original Action if the Signal is unchanged; HOLD if the Signal moved to NEUTRAL; never a stronger action than the original
ConvictionSTRONG only with HIGH confidence and no red flags; WEAK with any red flag

Example: original BULLISH 7.4, Confidence Score 62 (MEDIUM) → Score 5.0 + 2.4 × 0.62 = 6.5 → BULLISH / MEDIUM / original action / MODERATE.

In the output, the box below is followed immediately by this machine-readable footer — the same values as the box, filled in (no placeholders, no comments), so tools can read the result without parsing the box:

{"investskill": "1", "skill": "result-validator", "ticker": "<TICKER, or null for a portfolio / market-wide view>", "as_of": "<YYYY-MM-DD from Data & Sources>", "signal": "<BULLISH | NEUTRAL | BEARISH>", "confidence": "<HIGH | MEDIUM | LOW>", "horizon": "<SHORT-TERM | MEDIUM-TERM | LONG-TERM>", "score": 0.0, "action": "<BUY | HOLD | SELL>", "conviction": "<STRONG | MODERATE | WEAK>", "retrieval": "<pasted by user | web/tool retrieval | model memory>", "sources": ["<each primary source, dated>"]}
╔══════════════════════════════════════════════╗
║              INVESTMENT SIGNAL               ║
╠══════════════════════════════════════════════╣
║ Signal:      BULLISH / NEUTRAL / BEARISH     ║
║ Confidence:  HIGH / MEDIUM / LOW             ║
║ Horizon:     SHORT / MEDIUM / LONG-TERM      ║
║ Score:       X.X / 10                        ║
╠══════════════════════════════════════════════╣
║ Action:      BUY / HOLD / SELL               ║
║ Conviction:  STRONG / MODERATE / WEAK        ║
╚══════════════════════════════════════════════╝

Score Guide: 8.0–10.0 Strongly Bullish | 6.0–7.9 Moderately Bullish | 4.0–5.9 Neutral | 2.0–3.9 Moderately Bearish | 0.0–1.9 Strongly Bearish Confidence: HIGH (strong data, clear signals) | MEDIUM (mixed signals) | LOW (limited data, conflicting signals) Horizon: SHORT-TERM (1 week–3 months) | MEDIUM-TERM (3 months–1 year) | LONG-TERM (1+ years)

Disclaimer: Educational analysis only. Not financial advice.

Risk Stress Test — How Much Could This Portfolio Lose?

⚠️ Data Verification — Do This Before Any Analysis

Before running any analysis, always retrieve the latest market data for the ticker:

  1. Fetch current price — use web search or ask the user for the live price of every holding, so weights are computed on today's values, not on cost basis. Never assume a price from training data.
  2. Confirm the risk inputs — for each holding: 1-year beta to SPY, annualized volatility (σ), 30-day average daily volume (ADV) in shares and dollars, sector, and the current weight. These are the raw material of every number below; if they are missing, ask for them or state the proxy used (sector ETF beta / σ) next to the figure.
  3. State your data source — fill in the Data & Sources header (next section) so the origin, as-of date, retrieval path, and confidence of every figure are explicit at the top of the output.
  4. Flag stale data explicitly — if live data is unavailable, display this warning before proceeding:

⚠️ Live data unavailable. The following analysis uses training-data estimates which may be significantly out of date. Verify all prices and metrics before making any decisions.

Never silently substitute training-data estimates for current prices. When in doubt, ask the user to paste the latest quote.


📋 Data & Sources Header — Open Every Output With It

The first thing in the output is this provenance block, filled in — never left as placeholders. It is the standard documented on the Data & Accuracy page and the first thing result-validator looks for:

Data & Sources
  As of:      <date the figures represent, e.g. 2026-06-30>
  Source:     <primary docs — SEC EDGAR 10-K/10-Q, company IR, FRED, exchange data …>
  Retrieval:  <pasted by user | web/tool retrieval | model memory>
  Confidence: <HIGH | MEDIUM | LOW>
  • Retrieval: model memory must be paired with Confidence: LOW — memory is a placeholder until confirmed against a primary source.
  • Mixed sources: list each with its own as-of date rather than blending them.
  • Sources disagree on the same figure (a closing price, a deal term): prefer the primary source (SEC filing, company release, exchange data) over an aggregator, then the most recent dated one. Show both values with their sources and dates, say which you used and why, and never average them.
  • Data the user pasted is reported as pasted by user; do not upgrade its confidence beyond what the user's own source supports.

🌐 Output Language — --lang zh-TW

English is the default. With --lang zh-TW — or when the user writes in Traditional Chinese and asks for a Chinese answer — write the whole analysis in Traditional Chinese (繁體中文, Taiwan usage), keeping it machine-checkable:

  • Translate every heading, narrative paragraph, table header and bullet. Give each metric's English name in brackets the first time it appears — 自由現金流(Free Cash Flow, FCF) — so the reader can match it to the glossary and to English analyses.
  • Keep in English tickers, figures and units, SEC form and item names (10-K Item 7, Form 4), the four Data & Sources field labels (As of, Source, Retrieval, Confidence — the Retrieval value may be written in Chinese, e.g. 使用者提供), and the Investment Signal box and JSON footer exactly as specified — field names and the BULLISH / NEUTRAL / BEARISH, HIGH / MEDIUM / LOW, BUY / HOLD / SELL, STRONG / MODERATE / WEAK values. Tools parse them.
  • Title the invalidation section 論點失效條件(Thesis Invalidation), and close with the disclaimer in Chinese: 免責聲明: 僅供教育用途,非投資建議。
  • Use Taiwan financial terms: 殖利率、本益比、營業利益率、自由現金流、聯準會、選擇權、買權/賣權、除息日、預扣稅.

Overview

Every other framework in the catalog asks whether a holding is good. This one asks the question that decides whether the investor is still around to find out: how much can the whole book lose, how fast, and can it be sold when it has to be?

portfolio-review covers allocation, drift, and concentration. economics-analysis names the regime. Neither converts a list of holdings into a dollar figure for a bad month, replays the portfolio through the crashes that actually happened, or tells the investor how many days it would take to get out. This skill does exactly that and nothing else — it is a risk report, not a recommendation. Its verdict is a Risk Budget Score: how far the portfolio's measured risk sits inside (or outside) the drawdown the investor has said they can tolerate.

What an assistant can and cannot compute here. A language model cannot recall a covariance matrix for a specific set of tickers, and must not pretend to. The skill therefore works from inputs the user pastes or the assistant retrieves — beta, σ, ADV, weights — and shows every formula it applies, so the arithmetic can be checked line by line. Historical scenario magnitudes are approximate index-level figures and are labelled as such; the reader confirms the ones that matter.

Pairs with: portfolio-review (what to hold), economics-analysis (which regime the stress is most likely to come from), position-ladder (the share-count ceiling this report may lower), thesis-tracker (whether a drawdown is a thesis event or just volatility), and tax-lens (what a forced de-risking would cost after tax).


1. Inputs

InputRequiredDefault if unstated
Holdings with weights (or share counts + prices)✅—
Per-holding beta (1Y vs. SPY), annualized σ, 30-day ADV✅ (paste, retrieve, or state the proxy)sector-ETF beta / σ, flagged proxy
Total portfolio value✅ (for dollar figures)express everything in %
Max-drawdown tolerance (the risk budget)recommendedpropose one from the investor profile: 15 % conservative · 25 % balanced · 40 % aggressive
Horizonoptional1 day and 1 month
BenchmarkoptionalSPY
Cash / bonds / hedges (puts, inverse ETFs)optionalassumed zero — say so

Ask for missing risk inputs rather than inventing them. If the user cannot supply β or σ for a name, use the sector ETF's figures and mark every derived number (proxy).


2. Framework

Phase 1 — Exposure map (what the book actually is)

Weights first, then what they load onto:

Weight_i        = Value_i / Portfolio Value
β_p (net beta)  = Σ Weight_i × β_i
$-beta to SPY   = β_p × Portfolio Value          (dollar move for a 1.00 % SPY move × 100)
Sector weights  = Σ Weight_i by GICS sector      (flag any sector > 30 %)
Top-3 weight    = sum of the three largest positions
PositionWeightβσ (ann.)SectorWeight × βADV ($)
…
Portfolio100 %β_p(Phase 4)

Say plainly what the book is: "this is a 1.3-beta, 62 %-technology portfolio" is more useful than any statistic that follows.

Phase 2 — Historical scenario replay

Replay four regimes. Measure every holding over the scenario's common date window (the dates in the first column, index peak to index trough) — not each holding's own peak-to-trough, which would combine losses that never happened at the same time. Where a holding existed through the episode, use its actual return over that window; where it did not (or the data is not available), use β_i × index move or the sector ETF's return over the same window, and mark the row proxy. For the 2025 tariff shock use the index window below (the 2025-02-19 record close to the 2025-04-08 closing low); if the user prefers the announcement-to-trough window (2025-04-02 → 2025-04-08), say so and use it consistently. Index magnitudes below are approximate figures for orientation — the user confirms any that drive the conclusion.

ScenarioMarket move (approx.)Portfolio est. moveWorst positionNote
2008 GFC (Oct-07 → Mar-09)SPX ≈ −57 %, ~17 months; financials −80 %+, credit frozenΣ Weight_i × replay_iLong duration of decline matters as much as depth
March 2020 (Feb-19 → Mar-23)SPX ≈ −34 % in ~5 weeks; VIX > 80; everything sold at onceSpeed: correlations went to ~1; liquidity vanished for a fortnight
2022 rate shock (Jan → Oct)SPX ≈ −25 %; Nasdaq-100 ≈ −35 %; long-duration growth −60–80 %; bonds fell with stocksGrowth / unprofitable tech and 60/40 diversification were the casualties
2025 tariff shock (Feb-19 → Apr-8, 2025)SPX ≈ −19 % close to close, ≈ −10 % in the two sessions after the 2 April announcement; a sharp rebound began with the 9 April pauseImport-cost and China-revenue exposure, not beta, decided the losers
Portfolio scenario move = Σ Weight_i × Replay_i
   Replay_i = actual drawdown_i          if the holding traded through the episode
            = β_i × Index move            otherwise  (mark "proxy")
Dollar loss             = Portfolio scenario move × Portfolio Value

Report the worst single scenario in dollars, and whether it exceeds the risk budget. That comparison is the headline of the report.

Phase 3 — Parametric VaR and CVaR

σ_p        = portfolio σ from Phase 4 (use the "normal-correlation" figure)
VaR(α, t)  = z_α × σ_p × √t × Portfolio Value        z_95 = 1.645   z_99 = 2.326
CVaR(α, t) = (φ(z_α) / (1 − α)) × σ_p × √t × Value   CVaR_95 ≈ 2.063 σ√t · V   CVaR_99 ≈ 2.665 σ√t · V
t          = 1/252 for 1 day · 21/252 for 1 month   (σ_p is annualized)
HorizonVaR 95 %CVaR 95 %VaR 99 %CVaR 99 %
1 day$ / %
1 month$ / %

Say what this is and is not. Parametric VaR assumes normally distributed returns; equity tails are fatter, so realized 99 % losses come more often and run deeper than the table implies. Present VaR as the floor of what a bad day looks like, and the scenario replay in Phase 2 as the more honest picture of the tail. If the two disagree by more than 2×, say so and trust the scenario.

Phase 4 — Correlation spike: the diversification you think you have

Diversification is measured on calm days and disappears on the days it is needed. Compute portfolio σ twice:

σ_p²  = Σ_i Σ_j w_i w_j σ_i σ_j ρ_ij

Normal case : ρ_ij = the pasted / retrieved pairwise correlations
              (if unavailable: 0.6 within sector · 0.35 across sectors · 0.0 vs. cash — mark "assumed")
Crisis case : ρ_ij = 0.85 for every equity pair (0.95 within sector); bonds vs. equities 0.3
              (2022-style: not the −0.3 the 60/40 model assumes)

Diversification benefit = 1 − σ_p / Σ w_i σ_i        (report for both cases)
Portfolio σ (ann.)1-month VaR 99 %Diversification benefit
Normal correlations
Crisis correlations (ρ → 0.85)

The gap between the two rows is the amount of "diversification" the portfolio will lose exactly when it is needed. A portfolio whose crisis σ is more than 1.4× its normal σ is diversified in name only.

Phase 5 — Factor sensitivities

Three exposures that historical beta hides. Score each Low / Medium / High with the weight of the portfolio that carries it and the direction of the hit:

FactorWho gets hurtWho benefitsPortfolio exposureDirection of a +1 σ shock
Rates ↑ (10Y +100 bp)Long-duration growth (high P/S, distant cash flows), REITs, utilities, long bondsBanks (NIM), insurers, cash% of book in rate-sensitive namesest. % move
USD ↑ (DXY +5 %)Multinationals with > 40 % non-US revenue, commodity producers, EMDomestic small caps, importers% of book with heavy foreign revenue
Oil ↑ (WTI +25 %)Airlines, chemicals, consumer discretionary, transportsEnergy producers, oil services% energy vs. % oil-consumers

Use each company's segment / geographic disclosure (10-K) for revenue exposure; use the sector as a proxy only when stated.

Phase 6 — Liquidity: can it actually be sold?

Days to exit_i   = Shares held_i / (0.20 × ADV_i in shares)      (selling no more than 20 % of a day's volume)
Illiquid weight  = Σ Weight_i where Days to exit_i > 5
Portfolio exit   = the longest Days-to-exit among positions above 2 % weight
PositionSharesADV (shares)Days to exit @ 20 % ADVFlag
…> 5 days ⚠️ · > 20 days 🔴

A position that takes more than a month to exit at 20 % of ADV is a private holding with a ticker. Its drawdown in Phase 2 should be widened by an illiquidity haircut (10–25 %) and said so.

Phase 7 — Risk Budget Score (0–10)

The headline number. It measures how far inside the stated drawdown budget the portfolio sits — not whether the holdings are attractive.

ComponentWeight10 looks like0 looks like
Worst historical scenario vs. budget35 %Worst replay ≤ 60 % of the budgetWorst replay > 150 % of the budget
1-month CVaR 99 % vs. budget20 %CVaR ≤ 40 % of the budgetCVaR > budget
Correlation-spike resilience15 %Crisis σ ≤ 1.2× normal σCrisis σ > 1.6× normal σ
Concentration (top-3 weight, sector max)15 %Top-3 < 25 %, no sector > 30 %Top-3 > 50 % or one sector > 50 %
Liquidity10 %Every position exits in ≤ 2 days> 10 % of the book needs > 20 days
Factor tilt5 %No High exposure on any factorHigh on two or more factors, same direction
Risk Budget Score = Σ component score × weight        (each component scored 0–10, linearly between the anchors)

Map — the same bands as the standard Score Guide: ≥ 6.0 → risk within budget (signal BULLISH; 8.0+ comfortably so) · 4.0–5.9 → at the edge (NEUTRAL) · < 4.0 → over budget (BEARISH). If no budget was supplied, state the one you assumed and show how the score changes at the other two tolerance levels.


3. Output Format

  1. Data & Sources header (above) — including the source of every β / σ / ADV and which are proxies
  2. Headline — three lines: what the book is (β_p, largest tilts) · worst scenario in dollars vs. the budget · Risk Budget Score
  3. Exposure map (Phase 1 table)
  4. Scenario replay (Phase 2 table) with proxies marked
  5. VaR / CVaR table (Phase 3) with the normal-distribution caveat stated
  6. Correlation spike (Phase 4) — normal vs. crisis rows
  7. Factor sensitivities (Phase 5)
  8. Liquidity (Phase 6) with the longest exit named
  9. Risk Budget Score table (Phase 7) and the sensitivity to the budget assumed
  10. What would bring it inside budget — the two or three smallest changes (trim X to Y %, add Z % cash, replace A with a lower-β B) with the estimated effect on the worst scenario. Hand the tax cost of each to tax-lens; hand the how to position-ladder.
  11. Thesis Invalidation and the Investment Signal block

Example

User: Stress-test this portfolio against 2008 and 2022. $400k total. NVDA 22%, AVGO 15%,
      MSFT 15%, TSLA 12%, AMZN 10%, XOM 8%, JPM 8%, cash 10%. I can stomach a 25% drawdown.
      [pastes 1Y betas, σ, and ADV for each]

The assistant maps the book first: net β 1.28, 74% technology / consumer-tech, top-3 weight
52%, 10% cash. Scenario replay: 2008 ≈ −52% ($−208k, mostly proxy via β — none of the tech
names' 2008 paths are representative of today's businesses, and it says so); March 2020
≈ −41% in five weeks; 2022 rate shock ≈ −38% ($−152k) with TSLA (−65%) and NVDA (−50%) as the
worst positions; 2025 tariff shock flagged as the regime where AVGO / NVDA China exposure, not
beta, sets the loss — magnitude left for the reader to confirm. Parametric 1-month VaR 99% is
$−61k, CVaR 99% $−70k — a third of the 2022 replay, so the assistant tells the user to trust the
replay, not the VaR. Crisis-correlation σ is 1.5× normal σ: the seven names diversify each
other on quiet days only. Liquidity is fine (every position exits in under a day). Factor tilt:
High on rates (long-duration growth), Medium on USD, oil roughly neutral (XOM vs. AMZN/TSLA).
Risk Budget Score 3.4 / 10 — every historical scenario except a mild one breaches the 25%
budget. Smallest fixes: trim TSLA 12 → 5% and NVDA 22 → 15% into cash (worst-scenario loss
falls to ≈ −29%), or accept the tail and restate the budget at 40%. Tax cost of the trims →
tax-lens; how to execute → position-ladder.

Notes

  • This is a report, not a trade list. The Risk Budget Score says whether the risk fits the budget; deciding what to do about it belongs to portfolio-review, position-ladder, and the investor.
  • Every number is only as good as the pasted inputs. State proxies, assumed correlations, and index magnitudes as such. If more than a third of the risk inputs are proxies, the header's Confidence is LOW regardless of source quality.
  • Scenario dates and magnitudes are approximate orientation figures for major indices; the user confirms any that decide the verdict, especially the recent 2025 regime.
  • Options and hedges: if the user holds puts or inverse products, model them as a negative-β sleeve with a stated hedge ratio, and note that hedges decay and roll — a hedge's protection is for the period it covers, not forever.
  • Rebalance the budget, not just the book. A user who cannot stomach the worst historical scenario for their current book has two honest choices: less risk, or a bigger stated budget. Present both.
  • Run economics-analysis first when the question is which of the four regimes is nearest; run thesis-tracker --update after a drawdown to separate a broken thesis from ordinary volatility.

Thesis Invalidation

After delivering the analysis signal, specify what would reverse it:

If signal is BULLISH (risk within budget) — the verdict breaks if:

  • Any position drifts above the concentration anchors in Phase 7 (top-3 > 50 % or one sector > 30 %) through price appreciation alone — the book re-rates itself without a trade
  • Realized 30-day correlations between the top holdings rise above 0.7 in calm markets — the crisis case is becoming the base case
  • economics-analysis flags a regime shift toward the scenario that produced this book's worst replay (late-cycle → recession, or a renewed rate-hiking path for a long-duration book)
  • ADV of any position above 5 % weight halves — the liquidity assumption no longer holds

If signal is BEARISH (over budget) — the verdict breaks if:

  • The proposed trims / cash raise are executed and the worst-scenario replay falls inside the budget (re-run to confirm the number, not the intention)
  • The user restates the drawdown budget with a written reason (horizon, income, other assets) — the score is relative to the budget, so an honest larger budget changes it
  • A hedge sleeve is added with a stated ratio and roll plan that brings the crisis-correlation σ inside the anchor

Re-run this analysis when:

  • Quarter-end, or any rebalance
  • Any single position moves ±25 % relative to the portfolio
  • VIX closes above 30, or the 10-year yield moves ±75 bp in a quarter
  • A contribution or withdrawal larger than 10 % of portfolio value
  • The investor's horizon, income needs, or stated drawdown budget change

Standard Signal Output

This skill measures whether the portfolio's risk fits the investor's budget, not whether its holdings will rise, so state the mapping: Signal: BULLISH means the measured risk is inside the stated drawdown budget with margin; NEUTRAL means it sits at the edge; BEARISH means the worst plausible scenario breaches it. Action: BUY here means "the current risk level is acceptable to hold" — never "buy more." Action: SELL means "reduce risk," not that any holding is a short. Read the direction of the individual holdings from stock-eval or bear-case, not from this block.

All analysis concludes with this standardized block:

In the output, the box below is followed immediately by this machine-readable footer — the same values as the box, filled in (no placeholders, no comments), so tools can read the result without parsing the box:

{"investskill": "1", "skill": "risk-stress-test", "ticker": "<TICKER, or null for a portfolio / market-wide view>", "as_of": "<YYYY-MM-DD from Data & Sources>", "signal": "<BULLISH | NEUTRAL | BEARISH>", "confidence": "<HIGH | MEDIUM | LOW>", "horizon": "<SHORT-TERM | MEDIUM-TERM | LONG-TERM>", "score": 0.0, "action": "<BUY | HOLD | SELL>", "conviction": "<STRONG | MODERATE | WEAK>", "retrieval": "<pasted by user | web/tool retrieval | model memory>", "sources": ["<each primary source, dated>"]}
╔══════════════════════════════════════════════╗
║              INVESTMENT SIGNAL               ║
╠══════════════════════════════════════════════╣
║ Signal:      BULLISH / NEUTRAL / BEARISH     ║
║ Confidence:  HIGH / MEDIUM / LOW             ║
║ Horizon:     SHORT / MEDIUM / LONG-TERM      ║
║ Score:       X.X / 10                        ║
╠══════════════════════════════════════════════╣
║ Action:      BUY / HOLD / SELL               ║
║ Conviction:  STRONG / MODERATE / WEAK        ║
╚══════════════════════════════════════════════╝

Score Guide: 8.0–10.0 Strongly Bullish | 6.0–7.9 Moderately Bullish | 4.0–5.9 Neutral | 2.0–3.9 Moderately Bearish | 0.0–1.9 Strongly Bearish Confidence: HIGH (strong data, clear signals) | MEDIUM (mixed signals) | LOW (limited data, conflicting signals) Horizon: SHORT-TERM (1 week–3 months) | MEDIUM-TERM (3 months–1 year) | LONG-TERM (1+ years)

Note: The Score above is the Risk Budget Score, mapped onto the standard scale for cross-skill comparability. It rates how far inside the stated drawdown budget the portfolio sits — not how attractive the holdings are. Confidence is LOW whenever more than a third of the β / σ / ADV inputs are proxies or assumed correlations were used.

Disclaimer: Educational analysis only. Not financial advice. VaR, scenario replays, and drawdown estimates are models under stated assumptions; realized losses can and do exceed them.

Sector Analysis

⚠️ Data Verification — Do This Before Any Analysis

Before running any analysis, always retrieve the latest market data for the sectors you compare:

  1. Fetch current levels — use web search or ask the user for the current price and 3-, 6- and 12-month performance of each sector ETF (XLK, XLF, …) and the S&P 500. Never assume a level from training data.
  2. Confirm key figures — each sector's forward P/E and earnings-revision trend, as applicable to this skill.
  3. State your data source — fill in the Data & Sources header (next section) so the origin, as-of date, retrieval path, and confidence of every figure are explicit at the top of the output.
  4. Flag stale data explicitly — if live data is unavailable, display this warning before proceeding:

⚠️ Live data unavailable. The following analysis uses training-data estimates which may be significantly out of date. Verify all prices and metrics before making any decisions.

Never silently substitute training-data estimates for current prices. When in doubt, ask the user to paste the latest quote.


📋 Data & Sources Header — Open Every Output With It

The first thing in the output is this provenance block, filled in — never left as placeholders. It is the standard documented on the Data & Accuracy page and the first thing result-validator looks for:

Data & Sources
  As of:      <date the figures represent, e.g. 2026-06-30>
  Source:     <primary docs — SEC EDGAR 10-K/10-Q, company IR, FRED, exchange data …>
  Retrieval:  <pasted by user | web/tool retrieval | model memory>
  Confidence: <HIGH | MEDIUM | LOW>
  • Retrieval: model memory must be paired with Confidence: LOW — memory is a placeholder until confirmed against a primary source.
  • Mixed sources: list each with its own as-of date rather than blending them.
  • Sources disagree on the same figure (a closing price, a deal term): prefer the primary source (SEC filing, company release, exchange data) over an aggregator, then the most recent dated one. Show both values with their sources and dates, say which you used and why, and never average them.
  • Data the user pasted is reported as pasted by user; do not upgrade its confidence beyond what the user's own source supports.

🌐 Output Language — --lang zh-TW

English is the default. With --lang zh-TW — or when the user writes in Traditional Chinese and asks for a Chinese answer — write the whole analysis in Traditional Chinese (繁體中文, Taiwan usage), keeping it machine-checkable:

  • Translate every heading, narrative paragraph, table header and bullet. Give each metric's English name in brackets the first time it appears — 自由現金流(Free Cash Flow, FCF) — so the reader can match it to the glossary and to English analyses.
  • Keep in English tickers, figures and units, SEC form and item names (10-K Item 7, Form 4), the four Data & Sources field labels (As of, Source, Retrieval, Confidence — the Retrieval value may be written in Chinese, e.g. 使用者提供), and the Investment Signal box and JSON footer exactly as specified — field names and the BULLISH / NEUTRAL / BEARISH, HIGH / MEDIUM / LOW, BUY / HOLD / SELL, STRONG / MODERATE / WEAK values. Tools parse them.
  • Title the invalidation section 論點失效條件(Thesis Invalidation), and close with the disclaimer in Chinese: 免責聲明: 僅供教育用途,非投資建議。
  • Use Taiwan financial terms: 殖利率、本益比、營業利益率、自由現金流、聯準會、選擇權、買權/賣權、除息日、預扣稅.

Analyze US market sectors and identify sector rotation opportunities based on economic cycles.

Sector Overview

Analyze the 11 S&P 500 sectors:

  1. Information Technology
  2. Healthcare
  3. Financials
  4. Consumer Discretionary
  5. Communication Services
  6. Industrials
  7. Consumer Staples
  8. Energy
  9. Utilities
  10. Real Estate
  11. Materials

Analysis Framework

  1. Sector Performance

    • Relative performance vs. S&P 500
    • Historical performance trends
    • Momentum and trend strength
    • Volatility analysis
  2. Economic Cycle Positioning

    • Early cycle: Financials, Technology, Industrials
    • Mid cycle: Industrials, Materials, Energy
    • Late cycle: Energy, Consumer Staples, Healthcare
    • Recession: Utilities, Consumer Staples, Healthcare
  3. Fundamental Metrics

    • Sector valuation (P/E, P/B vs. historical)
    • Earnings growth forecasts
    • Profit margin trends
    • Revenue growth outlook
  4. Macro Drivers

    • Interest rate sensitivity
    • Commodity price exposure
    • Economic growth correlation
    • Currency impact
    • Regulatory environment
  5. Technical Picture

    • Sector ETF chart patterns
    • Relative strength analysis
    • Support/resistance levels
    • Volume trends

Sector Rotation Strategy

  • Identify current economic cycle phase
  • Determine leading and lagging sectors
  • Assess rotation timing signals
  • Evaluate defensive vs. cyclical positioning
  • Consider factor tilts (value, growth, quality)

Top Holdings Analysis

  • Identify sector leaders and laggards
  • Analyze top 5-10 stocks per sector
  • Evaluate sector concentration risks
  • Find emerging opportunities

Sector Valuation Comparison Table

Use these ranges as historical context benchmarks. Always verify current figures against live data sources. Ranges reflect long-run averages across full market cycles; individual readings may deviate significantly in extremes.

SectorTypical P/E RangeTypical P/S RangeEV/EBITDA RangeDividend Yield RangeHistorical EPS Growth (10Y CAGR)
Information Technology22–40x4–10x15–30x0.5–1.5%12–18%
Healthcare16–28x1.5–4x12–20x1.5–2.5%8–12%
Financials10–16x2–4x (Price/Book 1–2x)8–14x2.0–3.5%7–11%
Consumer Discretionary18–35x0.8–2.5x10–20x0.5–1.5%10–15%
Communication Services15–28x2–5x10–18x0.5–2.0%6–12%
Industrials16–25x1–2.5x10–16x1.5–2.5%7–11%
Consumer Staples18–26x0.8–2x12–17x2.5–4.0%5–8%
Energy8–18x (volatile)0.5–1.5x5–12x3.0–5.5%3–8% (commodity-driven)
Utilities14–22x1.5–3x10–15x3.0–5.0%3–6%
Real Estate (REITs)30–60x (use P/FFO: 14–22x)5–12x15–25x3.5–6.0%4–8%
Materials12–22x0.8–2x8–14x2.0–3.5%5–10%

Notes:

  • P/E for Energy and Financials is highly cyclical — use normalized or through-cycle P/E.
  • REITs are best valued on Price/FFO (Funds From Operations) or EV/EBITDA, not standard P/E.
  • Dividend yield ranges shift with interest rate regimes; compare to prevailing 10Y Treasury for context.
  • P/S is most useful for early-stage growth sectors (Tech, Comm Services) where margins are expanding.

Sector Seasonality Calendar

Historical seasonal patterns based on decades of S&P 500 sector returns. These are tendencies, not guarantees — confirm with current macro backdrop and momentum before acting.

MonthHistorically Strong SectorsHistorically Weak SectorsKey Seasonal Driver
JanuaryFinancials, Small Caps, IndustrialsUtilities, Consumer Staples"January Effect," new year portfolio repositioning
FebruaryHealthcare, TechnologyEnergy, MaterialsEarnings season (Q4 reports), defensive rotation
MarchEnergy, Industrials, MaterialsReal Estate, UtilitiesSpring economic activity pickup, rate expectations reset
AprilConsumer Discretionary, TechnologyEnergyStrong earnings season (Q1), consumer spending uplift
MayConsumer Staples, Healthcare, UtilitiesIndustrials, Materials"Sell in May" defensive rotation begins
JuneEnergy (early summer driving demand)Consumer Discretionary, FinancialsFed meeting seasonality, summer slowdown
JulyTechnology, Consumer DiscretionaryEnergyQ2 earnings beats, summer consumer activity
AugustConsumer Staples, UtilitiesTechnology, IndustrialsThin liquidity, risk-off tendency, vacation season
SeptemberEnergyTechnology, Consumer DiscretionaryHistorically worst month for equities overall
OctoberFinancials, Industrials, TechnologyReal EstateQ3 earnings season begins, year-end setup
NovemberConsumer Discretionary, Technology, IndustrialsUtilities, EnergyPre-holiday retail strength, "Santa rally" setup
DecemberConsumer Discretionary, Consumer StaplesFinancialsHoliday spending, tax-loss harvesting, year-end flows

Cycle-Overlay Seasonality:

  • Recession entry: Utilities, Consumer Staples, Healthcare outperform regardless of month.
  • Early recovery: Financials, Technology, and Consumer Discretionary lead — often most pronounced in Q1/Q2 post-trough.
  • Commodity supercycles: Energy and Materials seasonal patterns amplify vs. normal years.

Sector Correlation Matrix

Use this matrix to understand diversification benefits and macro sensitivity when constructing multi-sector portfolios. Correlations are approximate long-run averages; they compress toward 1.0 during market stress.

Inter-Sector Correlation (approximate, long-run)

TechHealthFinDiscCommIndStaplesEnergyUtilREMats
Tech—Low+Low+Med+Med+Low+Low-Low-Low-Low-Low-
HealthcareLow+—Low-Low-Low+Low-Med+Low-Med+Low-Low-
FinancialsLow+Low-—Med+Low+Med+Low-Low+Low-Med+Low+
DiscMed+Low-Med+—Med+Med+Low-Low-Low-Low-Low+
Comm SvcsMed+Low+Low+Med+—Low+Low+Low-Low+Low-Low-
IndustrialsLow+Low-Med+Med+Low+—Low-Med+Low-Low-Med+
StaplesLow-Med+Low-Low-Low+Low-—Low-Med+Low+Low-
EnergyLow-Low-Low+Low-Low-Med+Low-—Low-Low-Med+
UtilitiesLow-Med+Low-Low-Low+Low-Med+Low-—Med+Low-
Real EstateLow-Low-Med+Low-Low-Low-Low+Low-Med+—Low-
MaterialsLow-Low-Low+Low+Low-Med+Low-Med+Low-Low-—

Key: Med+ = moderate positive correlation (0.4–0.7) | Low+ = low positive (0.1–0.4) | Low- = low negative or near-zero (-0.2–0.1)

Sector Sensitivity to Key Macro Variables

Macro VariableStrong PositiveModerate PositiveNeutralModerate NegativeStrong Negative
Rising interest ratesFinancialsEnergy, MaterialsIndustrials, TechConsumer Disc, Comm SvcsUtilities, Real Estate
Falling interest ratesUtilities, Real EstateTech, HealthcareStaplesFinancials—
Rising oil/energy pricesEnergyMaterials, IndustrialsHealthcareConsumer Disc, StaplesAirlines (Disc)
Falling oil pricesConsumer Disc, AirlinesStaples, TechFinancialsEnergyMaterials
USD strengtheningDomestic Staples, UtilitiesFinancialsHealthcareTech (exports), IndustrialsMaterials, Energy
USD weakeningTech (multinationals), MaterialsEnergy, IndustrialsHealthcareDomestic Utilities—
GDP accelerationIndustrials, Materials, EnergyTech, Financials, DiscComm Svcs—Utilities, Staples
Recession / GDP contractionUtilities, Staples, Healthcare—Comm SvcsFinancials, IndustrialsEnergy, Materials, Disc
Inflation risingEnergy, MaterialsReal EstateIndustrialsTech (multiple compression)Utilities, Staples
Inflation fallingUtilities, Real Estate, TechHealthcare, DiscFinancialsEnergyMaterials
Credit spread wideningUtilities, Staples, Healthcare—TechFinancials, Real EstateDisc, Industrials

Sector Momentum Scoring

Rank all 11 sectors on a 1–11 scale (1 = strongest, 11 = weakest) across four dimensions, then produce a composite rank. Update this scoring monthly or after significant macro events.

Scoring Dimensions

DimensionHow to ScoreData Source
3M Price ReturnRank sectors by 3-month total return vs. each otherBloomberg, ETF returns (XLK, XLV, etc.)
Earnings Revision Trend% of analysts raising forward EPS estimates (breadth)FactSet, Bloomberg consensus
Forward P/E vs. 10Y Historical AverageDiscount = high score; premium = low scoreFactSet, LSEG
Analyst SentimentNet buy ratings minus sell ratings as % of totalBloomberg, Refinitiv

Momentum Scorecard Template

Sector                  3M Return  EPS Revisions  Fwd P/E vs Hist  Analyst Sent.  Composite Rank
Information Technology  [1-11]     [1-11]         [1-11]           [1-11]         [avg rank]
Healthcare              [1-11]     [1-11]         [1-11]           [1-11]         [avg rank]
Financials              [1-11]     [1-11]         [1-11]           [1-11]         [avg rank]
Consumer Discretionary  [1-11]     [1-11]         [1-11]           [1-11]         [avg rank]
Communication Services  [1-11]     [1-11]         [1-11]           [1-11]         [avg rank]
Industrials             [1-11]     [1-11]         [1-11]           [1-11]         [avg rank]
Consumer Staples        [1-11]     [1-11]         [1-11]           [1-11]         [avg rank]
Energy                  [1-11]     [1-11]         [1-11]           [1-11]         [avg rank]
Utilities               [1-11]     [1-11]         [1-11]           [1-11]         [avg rank]
Real Estate             [1-11]     [1-11]         [1-11]           [1-11]         [avg rank]
Materials               [1-11]     [1-11]         [1-11]           [1-11]         [avg rank]

Composite Rank Interpretation

Composite RankAction
1–3Strong overweight — broad-based positive momentum
4–5Moderate overweight — mostly positive signals
6–7Neutral weight — mixed signals
8–9Underweight — mostly negative signals
10–11Avoid / underweight significantly — broad deterioration

Weighting Suggestion: Equal-weight all four dimensions as a starting point. Tilt to 40% price return + 30% EPS revisions + 20% valuation + 10% sentiment for a momentum-focused strategy.


Peer Benchmarking Within Sector

When analyzing a specific stock, compare it against its sector median to identify relative attractiveness. Use this template for every individual stock recommendation within a sector rotation context.

Single Stock vs. Sector Median Template

Stock: [TICKER] — [Company Name]
Sector: [GICS Sector]
Comparison Date: [Date] | Source: [FactSet / Bloomberg / Company Filings]

DIMENSION              STOCK VALUE    SECTOR MEDIAN    PREMIUM / DISCOUNT    SCORE (1-5)
─────────────────────────────────────────────────────────────────────────────────────────
VALUATION
  Forward P/E          [x.x]x         [x.x]x           [+/- x%]              [1-5]
  EV/EBITDA            [x.x]x         [x.x]x           [+/- x%]              [1-5]
  Price/Sales          [x.x]x         [x.x]x           [+/- x%]              [1-5]
  Price/Book           [x.x]x         [x.x]x           [+/- x%]              [1-5]
  Dividend Yield       [x.x]%         [x.x]%           [+/- x bps]           [1-5]

GROWTH
  Revenue Growth (TTM) [x.x]%         [x.x]%           [+/- x%]              [1-5]
  EPS Growth (FY est.) [x.x]%         [x.x]%           [+/- x%]              [1-5]
  Revenue Growth (3Y)  [x.x]%         [x.x]%           [+/- x%]              [1-5]

MARGINS & QUALITY
  Gross Margin         [x.x]%         [x.x]%           [+/- x bps]           [1-5]
  EBITDA Margin        [x.x]%         [x.x]%           [+/- x bps]           [1-5]
  Net Margin           [x.x]%         [x.x]%           [+/- x bps]           [1-5]
  ROIC                 [x.x]%         [x.x]%           [+/- x bps]           [1-5]
  ROE                  [x.x]%         [x.x]%           [+/- x bps]           [1-5]
  Debt/EBITDA          [x.x]x         [x.x]x           [+/- x%]              [1-5]
  FCF Yield            [x.x]%         [x.x]%           [+/- x bps]           [1-5]
─────────────────────────────────────────────────────────────────────────────────────────
COMPOSITE QUALITY SCORE                                                       [avg/5]

Quality Score Interpretation

ScoreInterpretation
4.5–5.0Sector leader — significant premium justified
3.5–4.4Above-sector-average quality — moderate premium justified
2.5–3.4In-line with sector — valuation should be at-market
1.5–2.4Below-sector quality — discount warranted
1.0–1.4Sector laggard — avoid unless deep value thesis exists

Scoring Convention: For valuation metrics, cheaper = higher score (a stock trading at a discount to peers on P/E scores 5, a premium scores 1). For growth, margins, and quality metrics, higher is better.


Sector-Specific Risk Factors

Each sector carries idiosyncratic risks beyond broad market beta. Always assess these in the context of the current macro and regulatory environment before establishing a position.

Information Technology

  • Regulatory / Antitrust Risk: Large-cap platforms (search, social, cloud) face ongoing EU Digital Markets Act enforcement, DOJ antitrust cases, and potential structural remedies. Headline risk can compress multiples even without earnings impact.
  • AI Disruption / Obsolescence Risk: Generative AI rapidly changes competitive positioning — incumbents may be disrupted faster than traditional product cycles. Evaluate whether a company is a beneficiary or a target.
  • Semiconductor Supply Chain & Export Controls: TSMC concentration, US-China export restrictions on advanced chips (EAR controls), and geopolitical risk in Taiwan can cause severe supply shocks.

Healthcare

  • FDA Approval Risk / Clinical Trial Binary Events: Drug pipeline stocks carry binary event risk at Phase 2/3 readouts and FDA PDUFA dates; a single rejection can cause 40–70% drawdowns.
  • Patent Cliff Risk: Major pharmaceuticals face loss of exclusivity (LOE) on blockbuster drugs — revenue can fall 80%+ within 2 years of generic entry; assess pipeline coverage vs. patent expiry schedule.
  • Drug Pricing / CMS Negotiation: IRA Medicare drug price negotiation and political pressure on list prices compress revenue visibility for large pharma and biotech.

Financials

  • Interest Rate Sensitivity (NIM Compression): Banks' net interest margins expand with rate hikes but compress when rates fall or the yield curve inverts; this is the single largest driver of bank earnings variability.
  • Credit Cycle Risk: Loan loss provisions surge in recessions; commercial real estate (CRE) exposure is a persistent concern for regional banks. Monitor non-performing loan (NPL) ratios closely.
  • Regulatory Capital Requirements (Basel III/IV): Evolving capital adequacy rules (CET1 requirements, stress test results) constrain buyback capacity and ROE generation for large banks.

Consumer Discretionary

  • Consumer Balance Sheet Health: This sector is most exposed to rising consumer debt, declining savings rates, and credit tightening — especially for big-ticket items (autos, appliances, travel).
  • Tariff and Import Cost Risk: Heavy reliance on offshore manufacturing (apparel, electronics, footwear) means tariff escalation directly compresses margins before pricing power can respond.
  • Secular Shift in Spending (Physical vs. Digital): Traditional retail faces ongoing structural displacement from e-commerce; under-differentiated brick-and-mortar operators face secular decline.

Communication Services

  • Streaming Profitability Inflection Risk: Media/streaming companies face pressure to convert subscriber growth to sustained free cash flow; content cost inflation and competition from tech giants compress margins.
  • Advertising Cyclicality: Digital ad revenue is highly correlated with GDP growth and corporate spending budgets — falls sharply in recessions (Google, Meta ad revenue dropped 15–25% in prior downturns).
  • Spectrum Allocation and Infrastructure Costs: Telecom operators face large, lumpy capex cycles tied to 5G and fiber buildouts, with uncertain return timelines and regulatory pricing constraints.

Industrials

  • Government Defense/Infrastructure Budget Dependency: Defense contractors and infrastructure-linked industrials are highly sensitive to Congressional appropriations, sequestration risk, and multi-year contract cancellations.
  • Supply Chain Disruption and Input Cost Inflation: Aerospace and industrial machinery have long production cycles — shortages in specialty materials (titanium, rare earth components) or labor can cause years-long delivery delays.
  • Labor Cost Pressure and Union Risk: Heavily unionized manufacturing sectors (aerospace, auto, rail) face periodic strike risk and multi-year wage escalation that can compress margins durably.

Consumer Staples

  • Private Label / Retailer Margin Squeeze: In cost-of-living crisis periods, consumers trade down to retailer own-brand products, reducing branded CPG companies' pricing power and volume.
  • Input Cost Volatility (Commodities, Packaging): Agricultural commodity inputs (wheat, corn, sugar, palm oil) and energy-intensive packaging are subject to price spikes that compress gross margins with a lag.
  • Emerging Market Currency and Political Risk: Many staples companies derive 30–50% of revenue from EM; local currency depreciation and political instability can materially impact reported earnings.

Energy

  • Commodity Price Cycle Risk: Oil and gas earnings are almost entirely driven by the WTI/Brent/Henry Hub price — a 20% oil price decline can eliminate 40–60% of sector earnings in a single quarter.
  • Energy Transition / Stranded Asset Risk: Long-duration upstream assets (offshore fields, oil sands) carry the risk of becoming stranded as renewable penetration accelerates and carbon pricing expands.
  • Geopolitical Supply Disruption: OPEC+ production decisions, Middle East conflict, Russian supply constraints, and US shale production responses create persistent supply uncertainty that makes earnings forecasting highly uncertain.

Utilities

  • Interest Rate / Bond Proxy Risk: Utilities are priced as bond proxies — rising 10Y Treasury yields directly compress valuations as yield-seeking capital rotates to risk-free alternatives; every 100bps rate rise compresses sector P/E by 1–2 turns historically.
  • Regulatory Rate Case Risk: Utility earnings are set by state/federal regulators through rate cases; adverse rulings can cap ROE and delay capital recovery for large infrastructure investments.
  • Renewable Build-Out Execution Risk: Ambitious clean energy transition capex (solar, wind, grid modernization) carries construction delay risk, cost overruns, and financing risk in a volatile rate environment.

Real Estate (REITs)

  • Interest Rate Sensitivity and Refinancing Risk: REITs use significant leverage; rising rates increase interest expense and cap rate expansion reduces property valuations. Floating-rate debt exposure is a critical variable.
  • Property-Type Secular Trends: Office REITs face structural demand destruction from hybrid work; retail REITs face e-commerce headwinds. Not all REIT sub-sectors face the same structural forces.
  • Credit Market Access: REITs must access capital markets regularly to fund growth; credit spread widening and bank lending tightness during stress periods can trap overleveraged operators.

Materials

  • Commodity Price Volatility: Earnings are almost entirely driven by copper, aluminum, gold, steel, or chemical feedstock prices — these are globally set and subject to large cyclical swings.
  • China Demand Dependency: China accounts for 50–60% of global demand for base metals; property sector weakness, infrastructure slowdown, or trade tensions in China have outsized effects on global Materials earnings.
  • Environmental Regulation and Mine Permitting: Mining and chemical companies face increasingly stringent environmental standards, permitting delays (often 5–10 years for new mines), and carbon pricing that raises operating costs durably.

Sector ETF Reference

SectorSPDR ETFAlternative
Information TechnologyXLKVGT, QQQ
HealthcareXLVVHT
FinancialsXLFVFH
Consumer DiscretionaryXLYVCR
Communication ServicesXLCVOX
IndustrialsXLIVIS
Consumer StaplesXLPVDC
EnergyXLEVDE
UtilitiesXLUVPU
Real EstateXLREVNQ
MaterialsXLBVAW

Output

Provide sector analysis with:

  • Current sector rankings and momentum
  • Economic cycle assessment
  • Sector rotation recommendations
  • Top stock picks within favored sectors
  • Sectors to underweight/avoid
  • Risk considerations by sector
  • Expected catalysts and timeframes
  • Implementation strategy (ETFs vs. individual stocks)

Keep recommendations aligned with macro outlook and risk management principles.

Thesis Invalidation

After delivering the analysis signal, specify what would reverse it:

If signal is BULLISH — thesis breaks if:

  • sector underperforms S&P 500 by >10% over 3 months AND rate regime turns unfavorable
  • [One or two more triggers drawn from this analysis's own drivers, each with a threshold]

If signal is BEARISH — thesis breaks if:

  • sector rotates into leadership AND sector P/E discount to S&P closes
  • [One or two more triggers drawn from this analysis's own drivers, each with a threshold]

Re-run this analysis when:

  • Next earnings release
  • Price moves ±15% from current level
  • 60 days have elapsed
  • Material news event (acquisition, leadership change, regulatory decision)

Standard Signal Output

All analysis concludes with this standardized block:

In the output, the box below is followed immediately by this machine-readable footer — the same values as the box, filled in (no placeholders, no comments), so tools can read the result without parsing the box:

{"investskill": "1", "skill": "sector-analysis", "ticker": "<TICKER, or null for a portfolio / market-wide view>", "as_of": "<YYYY-MM-DD from Data & Sources>", "signal": "<BULLISH | NEUTRAL | BEARISH>", "confidence": "<HIGH | MEDIUM | LOW>", "horizon": "<SHORT-TERM | MEDIUM-TERM | LONG-TERM>", "score": 0.0, "action": "<BUY | HOLD | SELL>", "conviction": "<STRONG | MODERATE | WEAK>", "retrieval": "<pasted by user | web/tool retrieval | model memory>", "sources": ["<each primary source, dated>"]}
╔══════════════════════════════════════════════╗
║              INVESTMENT SIGNAL               ║
╠══════════════════════════════════════════════╣
║ Signal:      BULLISH / NEUTRAL / BEARISH     ║
║ Confidence:  HIGH / MEDIUM / LOW             ║
║ Horizon:     SHORT / MEDIUM / LONG-TERM      ║
║ Score:       X.X / 10                        ║
╠══════════════════════════════════════════════╣
║ Action:      BUY / HOLD / SELL               ║
║ Conviction:  STRONG / MODERATE / WEAK        ║
╚══════════════════════════════════════════════╝

Score Guide: 8.0–10.0 Strongly Bullish | 6.0–7.9 Moderately Bullish | 4.0–5.9 Neutral | 2.0–3.9 Moderately Bearish | 0.0–1.9 Strongly Bearish Confidence: HIGH (strong data, clear signals) | MEDIUM (mixed signals) | LOW (limited data, conflicting signals) Horizon: SHORT-TERM (1 week–3 months) | MEDIUM-TERM (3 months–1 year) | LONG-TERM (1+ years)

Disclaimer: Educational analysis only. Not financial advice.

Short Interest Analysis

⚠️ Data Verification — Do This Before Any Analysis

Before running any analysis, always retrieve the latest market data for the ticker:

  1. Fetch current price — use web search or ask the user for the live price, 52-week range, and market cap. Never assume a price from training data.
  2. Confirm key figures — shares short and its settlement date (exchange reports twice a month), float, average daily volume (for days to cover), borrow fee / utilization if available, and the date of each figure.
  3. State your data source — fill in the Data & Sources header (next section) so the origin, as-of date, retrieval path, and confidence of every figure are explicit at the top of the output.
  4. Flag stale data explicitly — if live data is unavailable, display this warning before proceeding:

⚠️ Live data unavailable. The following analysis uses training-data estimates which may be significantly out of date. Verify all prices and metrics before making any decisions.

Never silently substitute training-data estimates for current prices. When in doubt, ask the user to paste the latest quote.


📋 Data & Sources Header — Open Every Output With It

The first thing in the output is this provenance block, filled in — never left as placeholders. It is the standard documented on the Data & Accuracy page and the first thing result-validator looks for:

Data & Sources
  As of:      <date the figures represent, e.g. 2026-06-30>
  Source:     <primary docs — SEC EDGAR 10-K/10-Q, company IR, FRED, exchange data …>
  Retrieval:  <pasted by user | web/tool retrieval | model memory>
  Confidence: <HIGH | MEDIUM | LOW>
  • Retrieval: model memory must be paired with Confidence: LOW — memory is a placeholder until confirmed against a primary source.
  • Mixed sources: list each with its own as-of date rather than blending them.
  • Sources disagree on the same figure (a closing price, a deal term): prefer the primary source (SEC filing, company release, exchange data) over an aggregator, then the most recent dated one. Show both values with their sources and dates, say which you used and why, and never average them.
  • Data the user pasted is reported as pasted by user; do not upgrade its confidence beyond what the user's own source supports.

🌐 Output Language — --lang zh-TW

English is the default. With --lang zh-TW — or when the user writes in Traditional Chinese and asks for a Chinese answer — write the whole analysis in Traditional Chinese (繁體中文, Taiwan usage), keeping it machine-checkable:

  • Translate every heading, narrative paragraph, table header and bullet. Give each metric's English name in brackets the first time it appears — 自由現金流(Free Cash Flow, FCF) — so the reader can match it to the glossary and to English analyses.
  • Keep in English tickers, figures and units, SEC form and item names (10-K Item 7, Form 4), the four Data & Sources field labels (As of, Source, Retrieval, Confidence — the Retrieval value may be written in Chinese, e.g. 使用者提供), and the Investment Signal box and JSON footer exactly as specified — field names and the BULLISH / NEUTRAL / BEARISH, HIGH / MEDIUM / LOW, BUY / HOLD / SELL, STRONG / MODERATE / WEAK values. Tools parse them.
  • Title the invalidation section 論點失效條件(Thesis Invalidation), and close with the disclaimer in Chinese: 免責聲明: 僅供教育用途,非投資建議。
  • Use Taiwan financial terms: 殖利率、本益比、營業利益率、自由現金流、聯準會、選擇權、買權/賣權、除息日、預扣稅.

Comprehensive analysis of short selling activity, squeeze potential, cost-of-borrow dynamics, and bearish positioning signals for US-listed stocks. Combines FINRA short interest data, options market signals, and technical context to assess directional risk.

Analysis Framework

1. Short Interest Overview

Measure the current level and trend of bearish positioning:

Core Short Interest Metrics

  • Short interest (shares): Total number of shares sold short and not yet covered or closed
  • Short float %: Short shares / Total float (shares available for public trading)
  • Short interest as % of shares outstanding: Short shares / Total shares outstanding (includes locked-up insider shares)
  • Float % is the more actionable metric — tighter supply amplifies squeeze dynamics

Short Float % Thresholds

Short Float %     Interpretation
<2%               Negligible — minimal bearish conviction
2-5%              Low — modest skepticism or hedging
5-10%             Moderate — meaningful bearish positioning
10-20%            Elevated — significant short thesis in market
20-30%            High — heavy bearish conviction, squeeze potential
>30%              Extreme — high risk environment (squeeze OR fundamental collapse)

Days to Cover (Short Interest Ratio)

  • Formula: Short interest (shares) / Average daily trading volume
  • Represents: how many trading days it would take all shorts to cover at normal volume
  • Also called: Short Interest Ratio (SIR)
DTC Range         Interpretation
<1 day            Very low — shorts can exit quickly
1-3 days          Low — manageable exit risk for shorts
3-5 days          Moderate — some covering friction
5-10 days         Elevated — meaningful covering difficulty
>10 days          High — forced covering creates significant price pressure
>20 days          Extreme — structural squeeze conditions present

Change in Short Interest Trend

  • Compare last 4 reporting periods (approximately 2 months)
  • Increasing short interest: bears building positions, growing conviction against stock
  • Decreasing short interest: short covering, thesis deterioration, or squeeze in progress
  • Spike in short interest: new negative catalyst or activist short report
  • Rapid decline in short interest: covering rally in progress, thesis proved wrong

2. Short Squeeze Potential Assessment

Squeeze Score (0-10 Composite)

Assign points based on the following conditions:

  • Short float > 20%: +3 points (high short interest is the core prerequisite)
  • Days to Cover > 10: +2 points (exit difficulty creates forced buying)
  • Recent positive catalyst (earnings beat, product launch, partnership, FDA approval): +2 points
  • Strong price momentum (price above 50-day moving average): +1 point
  • Limited share supply (small float < 50 million shares): +1 point
  • High borrow rate (annualized cost > 5%): +1 point (shorts paying carrying cost = time pressure)
Squeeze Score     Probability Assessment
0-2               Low — insufficient conditions for meaningful squeeze
3-4               Moderate — partial conditions met, monitor for catalysts
5-6               High — most conditions met, squeeze possible on catalyst
7-8               Very High — strong squeeze setup, catalyst likely to trigger
9-10              Extreme — textbook squeeze conditions, high probability event

Three Requirements for High-Probability Squeeze

All three must be present simultaneously:

  1. High short interest: Float% > 15% AND DTC > 7 days (supply-demand imbalance)
  2. Catalyst or forced covering trigger: positive news, options expiration forcing hedging, index inclusion, or institutional accumulation
  3. Limited downside floor: fundamental support (valuation floor, asset value) or strong technical support level that prevents shorts from adding

Historical Squeeze Pattern Reference

  • GameStop (GME) January 2021: 140% short float + retail coordination + options gamma cascade
  • Volkswagen (VW) October 2008: Porsche cornered supply, DTC extreme, short float near 13% of total shares outstanding
  • Common precursors: borrow rate spikes sharply before price spike, DTC > 10 sustained for weeks, options call OI builds in OTM strikes
  • Key lesson: squeezes are self-reinforcing — covering buying creates price increase which triggers more covering

3. Borrow Rate and Cost of Short

Borrow Rate Tiers

Annualized Rate    Classification    Interpretation
<1%               General Collateral  Easy to borrow — no friction on shorts
1-5%              Easy to Borrow      Low cost, ample supply available
5-15%             Hard to Borrow      Elevated cost, supply becoming scarce
15-30%            Very Hard           Significant carry cost, time pressure on shorts
>30%              Extreme / Recall    Near impossible to maintain at sustainable cost

Borrow Rate Impact Analysis

  • At 20% annualized borrow rate: a short position costs shorts approximately 20 cents per $1 of position per year
  • High borrow rates create urgency to cover — especially if stock trades sideways or up
  • Borrow availability changes: easy to borrow → hard to borrow transition signals supply tightening
  • Recall risk: Prime brokers can recall shares on loan with short notice, forcing mandatory covering
  • Borrow rate spike (rapid increase in 1-2 weeks) is a leading indicator of pending squeeze

Implications of Borrow Dynamics

  • High borrow cost + high short interest = unstable equilibrium
  • Time works against short sellers in high-borrow-rate environments
  • Tracking borrow rate trend over 30-60 days is more valuable than single-point reading

4. Bearish Thesis Evaluation

Understand why short sellers are positioned against the company:

Common Short Thesis Categories

  • Accounting concerns: Revenue recognition irregularities, off-balance-sheet liabilities, undisclosed related-party transactions (Hindenburg Research, Muddy Waters Research methodologies)
  • Business model deterioration: Loss of competitive advantage, customer churn, unit economics breakdown
  • Competition threats: New entrant disrupting market, margin compression from rivals
  • Regulatory or legal risk: Pending investigations, product liability, antitrust exposure
  • Valuation excess: Momentum short — stock priced for perfection with no margin of safety (SPAC shorts, meme stocks)
  • Structural decline: Industry secular headwinds, obsolescence risk, debt spiral

Short Seller Credibility Assessment

  • Known activist short sellers (Muddy Waters, Citron, Gotham City, Kerrisdale; Hindenburg Research closed in January 2025, but its past reports still move stocks when revisited): publish detailed reports, track record publicly known
  • Hedge fund shorts: not visible in 13F filings, which report long positions only. US short positions surface only in aggregate (exchange short-interest reports twice a month; the SEC's Rule 13f-2 / Form SHO aggregate reporting once in force) or through the fund's own letters; in Europe, net shorts above 0.5% of share capital are published by name. Usually fundamental/valuation-based
  • Anonymous internet posts: low credibility, but sometimes contain legitimate leads
  • Evaluate track record: success rate of prior short thesis for identified short seller

Counterarguments to Short Thesis

  • What data would disprove the bearish case?
  • Has management addressed the concerns publicly?
  • Are there independent analysts or auditors refuting the claims?
  • Is there a fundamental value floor (book value, sum-of-parts) that limits downside?
  • Has the stock already priced in the worst-case scenario?

5. Options Market Signal Integration

Read options positioning as a forward-looking indicator of expected moves:

Put/Call Ratio Analysis

  • Put/Call open interest ratio > 1.5: heavy hedging or speculative bearish bet
  • Put/Call volume ratio spike: institutional protection buying (precedes downside)
  • Call/Put ratio spike: aggressive bullish buying or short hedge activity
  • Compare current ratio vs. 30/60/90-day average to identify anomalies

Implied Volatility Skew

  • Negative skew (normal for stocks): OTM put IV > OTM call IV — market pricing asymmetric downside
  • Skew steepening: increasing fear of downside, shorts adding OTM put protection
  • Skew flattening or reversal: call buying exceeds put buying — squeeze setup or bullish catalyst anticipated
  • Track 25-delta put IV minus 25-delta call IV as skew measure

Large Options Activity Signals

  • Large OTM put purchases (protective hedges from longs or speculative shorts)
  • Heavy OTM call buying near key resistance: anticipation of short squeeze catalyst
  • Gamma exposure at strikes: dealer hedging requirements create mechanical price pressure at specific levels
  • Unusual options activity 1-5 days before price move: informed positioning

6. Technical Analysis of Short Positions

Short Seller Cost Basis Estimation

  • Average price of shares shorted (based on short interest trend vs. stock price timeline)
  • Are shorts currently at a profit (stock below average short price) or a loss (stock above)?
  • Underwater shorts = trapped shorts = higher squeeze probability as losses mount

Key Price Levels for Short Positioning

  • Resistance zones where shorts typically initiate positions
  • Prior failed breakout levels (shorts cluster above failed resistance)
  • 52-week high: shorts often add above if they believe stock is overvalued
  • Volume-weighted average price of recent short interest build

Short Covering Patterns in Price Action

  • Sharp V-reversal on elevated volume with no obvious news: forced covering
  • Gap-up openings held and extended: short sellers unable to lean on stock
  • Failure to break below support despite negative sentiment: shorts losing control
  • Morning spike followed by day-long grind higher: staggered covering through session

Support Levels Affecting Covering Triggers

  • Key technical support: breakdown below = shorts add, hold = shorts must cover
  • 200-day moving average: institutional reference level where shorts reconsider
  • Prior consolidation zones: base building = shorts squeezed out on breakout

7. Reporting Schedule and Data Lag

FINRA Short Interest Reporting Calendar

  • Short interest is reported as of the 15th and last business day of each month
  • Data published approximately 4-5 business days after the settlement date
  • Result: up to a 2-week data lag between actual positioning and publicly available data
  • Critical caveat: short interest data reflects past positioning, not current

Exchange-Reported Short Interest

  • NYSE and NASDAQ publish their own short interest data on similar schedules
  • Cross-reference exchange data with FINRA for completeness
  • Real-time short volume (vs. short interest) available daily from FINRA and exchanges

Using Short Volume Data for Fresher Signal

  • Daily short volume (available same day via FINRA Rule 4560): not equal to short interest
  • Short volume / total volume ratio: measures daily bearish activity direction
  • Short volume ratio > 50% sustained over multiple days: active selling pressure
  • Limitation: covers only on-exchange transactions, misses OTC activity

8. Risk Considerations

Risks for Short Sellers

  • Unlimited loss potential: no cap on how high a stock can rise
  • Borrow recall: prime broker can force position closure on short notice
  • Dividend obligation: short sellers must pay dividends to the shares they borrowed
  • Regulatory restrictions: short sale restriction (SSR) rule triggers when stock falls 10% in a day, restricting short sales
  • Crowded short: many shorts in same name = all exit simultaneously on negative news = violent squeeze

Risks for Long Traders Playing Squeeze

  • Crowded momentum trade: late entrants face no new buyers once squeeze exhausts
  • Fundamental deterioration: short thesis may be correct — stock declines post-squeeze
  • Entry timing risk: squeezes can reverse violently once covering is complete
  • Borrow constraints: difficulty exiting short at target because all brokers pulling borrow simultaneously
  • Regulatory risk: market manipulation scrutiny for coordinated squeeze activity

Position Sizing Guidelines

  • Short squeeze plays: high risk/high reward, size accordingly (1-3% of portfolio maximum)
  • Risk is binary: squeeze fully materializes or fundamental thesis plays out
  • Define exit criteria before entry — both upside target and stop-loss level

9. Data Sources

Short Interest Data

  • FINRA Short Interest Data: finra.org/investors/short-sale-volume-data (official, free, twice-monthly)
  • NYSE Short Interest Reports: NYSE-listed stocks, semi-monthly
  • NASDAQ Short Interest Reports: NASDAQ-listed stocks, semi-monthly
  • Shortvolume.com: Free daily short volume data by ticker

Screeners and Analytics

  • Finviz: Short float % in stock screener (sortable, free)
  • Market Chameleon: Short interest trend charts, borrow rate tracking
  • Iborrowdesk.com: Free near-real-time borrow rate data from Interactive Brokers and TD Ameritrade

Premium Short Data

  • S3 Partners: Professional-grade short analytics, real-time borrow rates, predictive short interest
  • Interactive Brokers Stock Loan Availability: Borrow rate and availability for account holders
  • Bloomberg Terminal: SRFD function for comprehensive short interest data

Input Formats

Format 1: Single Stock Analysis

User: /short-interest GME

Claude retrieves current short float, DTC, borrow rate, and computes Squeeze Score

Format 2: Squeeze Potential Scan

User: /short-interest --scan squeeze-potential

Claude screens for stocks meeting high Squeeze Score criteria (short float >20%, DTC >10, catalyst present)

Format 3: Sector Short Analysis

User: /short-interest --sector technology

Claude identifies most shorted technology stocks and evaluates comparative squeeze potential

Output

Provide a comprehensive short interest analysis report with the following sections:

1. Short Interest Summary

Ticker:                  [TICKER]
Short Float %:           XX.X%       (Threshold: >20% = High)
Short Interest (shares): XX,XXX,XXX
Days to Cover (DTC):     X.X days    (Threshold: >10 = Elevated)
Borrow Rate:             X.X%        (Classification: Easy/Hard/Extreme)
Short Interest Trend:    Increasing / Stable / Decreasing
Last Reporting Date:     [Date]      (Note: ~2-week lag)

2. Squeeze Score Breakdown

Component                     Condition          Points
Short float > 20%             YES / NO           X/3
DTC > 10 days                 YES / NO           X/2
Recent positive catalyst       YES / NO           X/2
Price above 50-day MA         YES / NO           X/1
Float < 50M shares            YES / NO           X/1
Borrow rate > 5%              YES / NO           X/1
                              TOTAL SCORE:       X/10

Squeeze Probability: Low / Moderate / High / Very High / Extreme

3. Short Thesis Summary

  • Primary bear thesis (1-3 sentence summary of why shorts are positioned here)
  • Source of short thesis (activist report, valuation, fundamental deterioration)
  • Short seller credibility assessment

4. Counter-Thesis

  • Key arguments that could prove shorts wrong
  • Fundamental floor (book value, earnings power)
  • Upcoming catalysts that could force covering

5. Options Market Context

  • Put/Call ratio vs. 30-day average
  • IV skew assessment (steep put skew vs. flat)
  • Any unusual options activity flagged

6. Technical Setup

  • Is stock above or below 50/200-day moving average?
  • Key support and resistance levels
  • Is price action consistent with squeeze initiation or continued decline?

7. Risk/Reward Assessment

  • For long (squeeze play): upside target, stop-loss, probability-weighted expected return
  • For short (bearish thesis): downside target, covering trigger, maximum loss scenario

8. Monitoring Triggers

  • Short interest change threshold that would upgrade or downgrade squeeze probability
  • Catalyst calendar (earnings date, product event, regulatory decision)
  • Borrow rate level that signals imminent covering pressure
  • Price level that would technically confirm squeeze initiation

Thesis Invalidation

After delivering the analysis signal, specify what would reverse it:

If signal is BULLISH — thesis breaks if:

  • short float rises above 20% AND days-to-cover increases >10 without a catalyst
  • [One or two more triggers drawn from this analysis's own drivers, each with a threshold]

If signal is BEARISH — thesis breaks if:

  • short float drops below 3% AND squeeze signals activate (borrow rate >50%)
  • [One or two more triggers drawn from this analysis's own drivers, each with a threshold]

Re-run this analysis when:

  • Next earnings release
  • Price moves ±15% from current level
  • 60 days have elapsed
  • Material news event (acquisition, leadership change, regulatory decision)

Standard Signal Output

All analysis concludes with this standardized block:

In the output, the box below is followed immediately by this machine-readable footer — the same values as the box, filled in (no placeholders, no comments), so tools can read the result without parsing the box:

{"investskill": "1", "skill": "short-interest", "ticker": "<TICKER, or null for a portfolio / market-wide view>", "as_of": "<YYYY-MM-DD from Data & Sources>", "signal": "<BULLISH | NEUTRAL | BEARISH>", "confidence": "<HIGH | MEDIUM | LOW>", "horizon": "<SHORT-TERM | MEDIUM-TERM | LONG-TERM>", "score": 0.0, "action": "<BUY | HOLD | SELL>", "conviction": "<STRONG | MODERATE | WEAK>", "retrieval": "<pasted by user | web/tool retrieval | model memory>", "sources": ["<each primary source, dated>"]}
╔══════════════════════════════════════════════╗
║              INVESTMENT SIGNAL               ║
╠══════════════════════════════════════════════╣
║ Signal:      BULLISH / NEUTRAL / BEARISH     ║
║ Confidence:  HIGH / MEDIUM / LOW             ║
║ Horizon:     SHORT / MEDIUM / LONG-TERM      ║
║ Score:       X.X / 10                        ║
╠══════════════════════════════════════════════╣
║ Action:      BUY / HOLD / SELL               ║
║ Conviction:  STRONG / MODERATE / WEAK        ║
╚══════════════════════════════════════════════╝

Score Guide: 8.0–10.0 Strongly Bullish | 6.0–7.9 Moderately Bullish | 4.0–5.9 Neutral | 2.0–3.9 Moderately Bearish | 0.0–1.9 Strongly Bearish Confidence: HIGH (strong data, clear signals) | MEDIUM (mixed signals) | LOW (limited data, conflicting signals) Horizon: SHORT-TERM (1 week–3 months) | MEDIUM-TERM (3 months–1 year) | LONG-TERM (1+ years)

Disclaimer: Educational analysis only. Not financial advice.

US Stock Evaluation

⚠️ Data Verification — Do This Before Any Analysis

Before running any analysis, always retrieve the latest market data for the ticker:

  1. Fetch current price — use web search or ask the user for the live price, 52-week range, and market cap. Never assume a price from training data.
  2. Confirm key figures — the latest reported quarter and fiscal year (revenue, EPS, margins, FCF), current multiples (P/E, EV/EBITDA, P/FCF) computed from the confirmed price, net debt, and consensus estimates with their date.
  3. State your data source — fill in the Data & Sources header (next section) so the origin, as-of date, retrieval path, and confidence of every figure are explicit at the top of the output.
  4. Flag stale data explicitly — if live data is unavailable, display this warning before proceeding:

⚠️ Live data unavailable. The following analysis uses training-data estimates which may be significantly out of date. Verify all prices and metrics before making any decisions.

Never silently substitute training-data estimates for current prices. When in doubt, ask the user to paste the latest quote.


📋 Data & Sources Header — Open Every Output With It

The first thing in the output is this provenance block, filled in — never left as placeholders. It is the standard documented on the Data & Accuracy page and the first thing result-validator looks for:

Data & Sources
  As of:      <date the figures represent, e.g. 2026-06-30>
  Source:     <primary docs — SEC EDGAR 10-K/10-Q, company IR, FRED, exchange data …>
  Retrieval:  <pasted by user | web/tool retrieval | model memory>
  Confidence: <HIGH | MEDIUM | LOW>
  • Retrieval: model memory must be paired with Confidence: LOW — memory is a placeholder until confirmed against a primary source.
  • Mixed sources: list each with its own as-of date rather than blending them.
  • Sources disagree on the same figure (a closing price, a deal term): prefer the primary source (SEC filing, company release, exchange data) over an aggregator, then the most recent dated one. Show both values with their sources and dates, say which you used and why, and never average them.
  • Data the user pasted is reported as pasted by user; do not upgrade its confidence beyond what the user's own source supports.

🌐 Output Language — --lang zh-TW

English is the default. With --lang zh-TW — or when the user writes in Traditional Chinese and asks for a Chinese answer — write the whole analysis in Traditional Chinese (繁體中文, Taiwan usage), keeping it machine-checkable:

  • Translate every heading, narrative paragraph, table header and bullet. Give each metric's English name in brackets the first time it appears — 自由現金流(Free Cash Flow, FCF) — so the reader can match it to the glossary and to English analyses.
  • Keep in English tickers, figures and units, SEC form and item names (10-K Item 7, Form 4), the four Data & Sources field labels (As of, Source, Retrieval, Confidence — the Retrieval value may be written in Chinese, e.g. 使用者提供), and the Investment Signal box and JSON footer exactly as specified — field names and the BULLISH / NEUTRAL / BEARISH, HIGH / MEDIUM / LOW, BUY / HOLD / SELL, STRONG / MODERATE / WEAK values. Tools parse them.
  • Title the invalidation section 論點失效條件(Thesis Invalidation), and close with the disclaimer in Chinese: 免責聲明: 僅供教育用途,非投資建議。
  • Use Taiwan financial terms: 殖利率、本益比、營業利益率、自由現金流、聯準會、選擇權、買權/賣權、除息日、預扣稅.

Perform comprehensive stock evaluation combining fundamental analysis, valuation modeling, quality scoring, and risk assessment to produce investment-grade conclusions.

Analysis Components

1. Company Overview

  • Business model and competitive advantages (moat assessment)
  • Market position, addressable market size, and industry trends
  • Revenue mix by segment and geographic exposure
  • Key products, services, and customer concentration
  • Competitive dynamics and threat of disruption

2. Financial Health

  • Revenue and earnings growth trends (3-year and 5-year CAGR)
  • Profit margins: gross margin, operating margin, net margin
  • Margin trends: expanding, stable, or compressing
  • Balance sheet strength: cash, total debt, net debt, book value
  • Liquidity: current ratio, quick ratio, cash conversion cycle
  • Cash flow analysis: operating cash flow, free cash flow, FCF yield
  • Capital expenditure requirements (maintenance vs. growth capex)
  • Working capital management efficiency

3. Valuation Metrics

MetricCurrent1-Year Ago5-Year AvgSector Avg
P/E (TTM)
P/E (Forward)
PEG Ratio
Price/Book
Price/Sales
EV/EBITDA
EV/FCF
Dividend Yield
Payout Ratio

4. Key Ratios

RatioCurrentIndustry AvgAssessment
Return on Equity (ROE)
Return on Assets (ROA)
Return on Inv. Capital
Gross Margin
Operating Margin
Net Margin
Current Ratio
Quick Ratio
Debt-to-Equity
Interest Coverage
Asset Turnover

5. Peer Comparison

  • Direct peer comparison within industry on key valuation multiples
  • Historical valuation trends vs. own 5-year history
  • Sector performance context and relative positioning
  • Market share trends vs. competitors

Deep Financial Statement Analysis

This section provides line-item rigor absorbed from dedicated fundamental analysis workflows. Use when a deeper statement-level breakdown is required.

Income Statement Line-Item Framework

Revenue Analysis

  • Break revenue into organic growth vs. acquisition-driven growth vs. FX tailwind/headwind
  • Identify top revenue segments by size and growth rate; flag any segment growing faster or slower than the consolidated total
  • Assess revenue quality: recurring subscription/contract revenue vs. transactional/one-time revenue
  • Operating leverage test: does revenue growth outpace operating expense growth? Calculate incremental operating margin = ΔOperating Income / ΔRevenue

Cost Structure Decomposition

  • COGS breakdown: raw materials, labor, overhead — track each as % of revenue over 5 years
  • SG&A as % of revenue: stable or creeping? Rising SG&A without revenue acceleration signals inefficiency
  • R&D intensity: absolute spend and % of revenue — context-dependent (biotech vs. consumer goods)
  • Non-recurring items to strip from normalized earnings: restructuring charges, asset write-downs, gain/loss on asset sales, one-time tax benefits, litigation settlements

Operating Leverage Analysis

  • Fixed vs. variable cost split (estimate from historical margin behavior at different revenue levels)
  • Degree of Operating Leverage (DOL) = % Change in Operating Income / % Change in Revenue
  • High DOL (>3x) amplifies both upside and downside from revenue swings
YearRevenue ($M)Revenue Growth %EBIT ($M)EBIT Growth %DOL
Y-4
Y-3
Y-2
Y-1
TTM

Working Capital Cycle Analysis

Working capital efficiency directly affects how much cash a business consumes or generates as it grows.

Key Working Capital Metrics

MetricFormulaCurrentY-1Y-2Trend
Days Sales Outstanding (DSO)Accounts Receivable / (Revenue / 365)
Days Inventory Outstanding (DIO)Inventory / (COGS / 365)
Days Payable Outstanding (DPO)Accounts Payable / (COGS / 365)
Cash Conversion Cycle (CCC)DSO + DIO − DPO
Net Working Capital ($M)Current Assets − Current Liabilities
NWC as % of RevenueNWC / Revenue

Cash Conversion Cycle Interpretation

  • CCC declining: Company is collecting faster, holding less inventory, and/or extending payables — favorable working capital dynamics, cash generation improves as revenue grows
  • CCC rising: Working capital is consuming more cash as the business scales — a hidden drag on FCF growth
  • Negative CCC (e.g., large retailers, subscription businesses): Customers pay before the company pays suppliers — a structural cash flow advantage

Working Capital as a Growth Funding Check

  • Incremental NWC per $1 of new revenue = ΔNWC / ΔRevenue
  • If this ratio exceeds ~15%, rapid revenue growth can strain liquidity even in a profitable business

DuPont Decomposition

DuPont analysis decomposes ROE into its constituent drivers to identify whether profitability, efficiency, or leverage is the primary return engine — and whether returns are sustainable.

3-Factor DuPont

ROE = Net Profit Margin × Asset Turnover × Equity Multiplier

ComponentFormulaCurrentY-1Y-2
Net Profit MarginNet Income / Revenue
Asset TurnoverRevenue / Average Total Assets
Equity MultiplierAverage Total Assets / Average Shareholders' Equity
ROEMargin × Turnover × Multiplier

5-Factor DuPont (Extended)

ROE = Tax Burden × Interest Burden × EBIT Margin × Asset Turnover × Equity Multiplier

ComponentFormulaCurrentY-1Y-2
Tax BurdenNet Income / Pre-Tax Income
Interest BurdenPre-Tax Income / EBIT
EBIT MarginEBIT / Revenue
Asset TurnoverRevenue / Average Total Assets
Equity MultiplierAverage Total Assets / Average Equity
ROEProduct of all five

DuPont Interpretation

  • ROE driven by high margin (e.g., luxury brands, software): High-quality, sustainable — least risky driver
  • ROE driven by high asset turnover (e.g., retailers, distributors): Efficient but thin — watch for margin erosion
  • ROE driven by high leverage (equity multiplier > 3x): Amplified returns but magnified downside — assess debt sustainability carefully
  • Declining ROE: diagnose which factor is deteriorating before drawing conclusions

Competitive Analysis — Porter's Five Forces

ForceIntensity (H/M/L)Key Evidence
Threat of New EntrantsCapital requirements, brand moats, regulatory barriers
Bargaining Power of SuppliersSupplier concentration, switching costs, input criticality
Bargaining Power of BuyersCustomer concentration, price sensitivity, alternatives
Threat of SubstitutesTechnology disruption, cross-industry competition
Competitive RivalryNumber of peers, market growth rate, differentiation

Overall Moat Assessment: Wide / Narrow / None

  • Wide moat: Durable competitive advantage expected to persist 20+ years
  • Narrow moat: Competitive advantage present but may erode within 10 years
  • No moat: Competing primarily on price or operational efficiency

Visualization Data Tables

When a visual or report output is needed, populate these tables for chart generation (compatible with /us-stock-analysis:report-generator):

Revenue & Earnings Growth

Year    Revenue ($M)    Revenue Growth %    Net Income ($M)    EPS ($)
FY-4    [value]         [%]                 [value]            [value]
FY-3    [value]         [%]                 [value]            [value]
FY-2    [value]         [%]                 [value]            [value]
FY-1    [value]         [%]                 [value]            [value]
FY0     [value]         [%]                 [value]            [value]    ← latest reported fiscal year (label with the actual years)

Profit Margin Trends

Year    Gross Margin %    Operating Margin %    Net Margin %    Industry Avg %
2020    [%]              [%]                   [%]             [%]
2021    [%]              [%]                   [%]             [%]
2022    [%]              [%]                   [%]             [%]
2023    [%]              [%]                   [%]             [%]
2024    [%]              [%]                   [%]             [%]

Balance Sheet Composition

Year    Current Assets ($M)    Fixed Assets ($M)    Intangibles ($M)    Total Assets ($M)
2020    [value]               [value]              [value]             [value]
2021    [value]               [value]              [value]             [value]
2022    [value]               [value]              [value]             [value]
2023    [value]               [value]              [value]             [value]
2024    [value]               [value]              [value]             [value]

Year    Current Liab ($M)    Long-term Debt ($M)    Equity ($M)    Total L+E ($M)
2020    [value]             [value]                [value]        [value]
2021    [value]             [value]                [value]        [value]
2022    [value]             [value]                [value]        [value]
2023    [value]             [value]                [value]        [value]
2024    [value]             [value]                [value]        [value]

Cash Flow Waterfall

Component                   Amount ($M)    Notes
Operating Cash Flow         [value]        Core business generation
Capital Expenditures        [value]        Investments in assets
Free Cash Flow             [value]        Available for distribution
Dividends                  [value]        Shareholder distribution
Share Buybacks             [value]        Share repurchases
M&A Activity               [value]        Acquisitions
Debt Repayment             [value]        Debt reduction
Net Cash Change            [value]        Bottom line impact

Valuation Multiples Comparison

Metric              Company    Industry Avg    5-Year Avg    Assessment
P/E Ratio           [value]    [value]         [value]       [Over/Under/Fair]
P/B Ratio           [value]    [value]         [value]       [Over/Under/Fair]
P/S Ratio           [value]    [value]         [value]       [Over/Under/Fair]
EV/EBITDA          [value]    [value]         [value]       [Over/Under/Fair]
PEG Ratio          [value]    [value]         [value]       [Over/Under/Fair]

Quality Scoring Framework

Piotroski F-Score (0–9)

The Piotroski F-Score measures financial strength and operating improvement across 9 binary criteria. Each criterion scores 1 (pass) or 0 (fail). Score of 8–9 is strong; 0–2 is weak.

Profitability Signals (4 criteria):

#CriterionFormulaPass ConditionScore
1ROA > 0Net Income / Total AssetsPositive ROA in current year0 or 1
2Operating Cash Flow > 0CFO from cash flow statementPositive CFO0 or 1
3Change in ROAROA(t) − ROA(t-1)ROA improved year-over-year0 or 1
4Accruals (quality)CFO / Total Assets > ROACash earnings exceed reported earnings0 or 1

Leverage, Liquidity & Source of Funds (3 criteria):

#CriterionFormulaPass ConditionScore
5Change in LeverageLong-term Debt / Avg AssetsLeverage decreased YoY0 or 1
6Change in LiquidityCurrent Ratio(t) vs (t-1)Current ratio improved YoY0 or 1
7No New Shares IssuedDiluted shares outstandingNo new equity issuance in past year0 or 1

Operating Efficiency (2 criteria):

#CriterionFormulaPass ConditionScore
8Change in Gross MarginGross Margin(t) vs (t-1)Gross margin expanded YoY0 or 1
9Change in Asset TurnoverRevenue / Total AssetsAsset turnover improved YoY0 or 1

F-Score Interpretation:

  • 8–9: Strong financial position — high-quality candidate
  • 5–7: Average quality — neutral
  • 0–2: Weak financial position — high risk of deterioration

Earnings Quality Score

  • Accruals Ratio: (Net Income − CFO) / Average Total Assets. Low or negative accruals indicate high earnings quality (cash-backed profits).
  • Cash Conversion Rate: CFO / Net Income. Ratio consistently above 1.0x is positive. Below 0.7x signals potential earnings inflation.
  • Non-Recurring Items: Identify and strip out restructuring charges, asset write-downs, gains on asset sales, and one-time tax benefits from normalized earnings.
  • Revenue Recognition Risk: Assess channel stuffing, bill-and-hold arrangements, and aggressive deferred revenue recognition.

ROIC / WACC Analysis

ROIC Calculation

ROIC = NOPAT / Invested Capital

  • NOPAT (Net Operating Profit After Tax) = EBIT × (1 − effective tax rate)
  • Invested Capital = Total Equity + Total Debt − Cash and Cash Equivalents
  • Alternatively: Invested Capital = Net PP&E + Net Working Capital + Goodwill + Other Long-term Operating Assets
ComponentValue
EBIT
Effective Tax Rate
NOPAT
Total Equity
Total Debt
Less: Cash
Invested Capital
ROIC

WACC Calculation

WACC = (E/V) × Re + (D/V) × Rd × (1 − T)

  • Cost of Equity (Re) via CAPM: Re = Rf + β × (Rm − Rf)
    • Rf: Risk-free rate (current 10-year Treasury yield)
    • β: Stock beta (5-year monthly vs. S&P 500)
    • (Rm − Rf): Equity risk premium (historical ~5–6%)
  • Cost of Debt (Rd): Weighted average interest rate on outstanding debt
  • Capital Structure Weights: E/V = market cap / (market cap + total debt), D/V = total debt / (market cap + total debt)
  • Tax Shield: Multiply Rd by (1 − marginal tax rate) to reflect interest deductibility
ComponentValue
Risk-Free Rate (Rf)
Beta (β)
Equity Risk Premium
Cost of Equity (Re)
Pre-Tax Cost of Debt
Tax Rate
After-Tax Cost of Debt
Equity Weight (E/V)
Debt Weight (D/V)
WACC

Economic Value Added (EVA)

EVA = (ROIC − WACC) × Invested Capital

  • ROIC > WACC: Company is creating economic value — every dollar invested earns more than its cost. Strong positive signal.
  • ROIC = WACC: Company is breaking even economically — no value creation or destruction.
  • ROIC < WACC: Company is destroying shareholder value — capital is deployed below its opportunity cost. Significant warning sign.

ROIC vs. WACC spread trend (expanding = improving value creation, compressing = deteriorating):

YearROICWACCSpreadEVA
Current
-1 Year
-2 Year
-3 Year

DCF Framework

A quick DCF, not the valuation of record. This section gives a fast intrinsic-value range to place the stock in context. For the full model — WACC decomposition, three scenarios, sensitivity tables and comparables — run stock-valuation, and when the two disagree, use its number. Both use the same Gordon-growth terminal value; this quick version cross-checks with an EV/EBITDA exit multiple, stock-valuation with EV/FCF, so expect small differences.

Step-by-Step DCF Methodology

Step 1 — Project Free Cash Flow (Years 1–10)

  • Start with current year base FCF = Operating Cash Flow − Maintenance Capex
  • Apply revenue growth rate assumptions (differentiate Phase 1: high growth, Phase 2: fade to stable)
  • Apply operating margin assumptions (expanding/stable/compressing)
  • Deduct taxes and changes in net working capital
  • FCF = NOPAT + D&A − Change in Working Capital − Capex

Step 2 — Terminal Value Calculation

  • Gordon Growth Model: TV = FCF(Year 10) × (1 + g) / (WACC − g)
  • Exit Multiple Method: TV = EBITDA(Year 10) × Terminal EV/EBITDA multiple
  • Terminal growth rate (g): typically 2–3% (in line with long-run nominal GDP growth)
  • Cross-check both methods for reasonableness

Step 3 — Discount to Present Value

  • Discount each year's FCF: PV(FCFt) = FCFt / (1 + WACC)^t
  • Discount terminal value: PV(TV) = TV / (1 + WACC)^10
  • Enterprise Value = Sum of all discounted FCFs + PV(Terminal Value)
  • Equity Value = Enterprise Value − Net Debt
  • Intrinsic Value Per Share = Equity Value / Diluted Shares Outstanding

Key Assumptions

AssumptionBase CaseBull CaseBear Case
Revenue Growth Yr 1–5
Revenue Growth Yr 6–10
Operating Margin
Tax Rate
Capex (% Revenue)
WACC
Terminal Growth Rate
Terminal EV/EBITDA

Sensitivity Table

Intrinsic value per share at various WACC and terminal growth rate combinations:

WACC \ Terminal Growth1.5%2.0%2.5%3.0%3.5%
7%
8%
9%
10%
11%

Margin of Safety Assessment

  • Intrinsic Value (Base Case):
  • Current Market Price:
  • Premium / (Discount) to Intrinsic Value:
  • Margin of Safety: (Intrinsic Value − Market Price) / Intrinsic Value × 100%
    • 30% discount: Significant margin of safety — compelling value

    • 10–30% discount: Moderate margin of safety — attractive
    • 0–10% discount: Limited margin of safety — fairly valued
    • Trading at premium: No margin of safety — risk of capital loss if assumptions miss

Management Quality Assessment

Capital Allocation Track Record

  • ROIC Trend: Is management generating returns above cost of capital consistently? Review 5-year ROIC history.
  • Acquisition History: Evaluate past M&A for value creation. Did acquired assets earn above WACC? Were goodwill impairments taken post-acquisition?
  • Buyback Timing: Did management repurchase shares below intrinsic value? Assess buyback price vs. subsequent performance.
  • Dividend Policy: Sustainable payout ratio? Dividend growth track record. Balance between dividends, buybacks, and reinvestment.
  • Capex Discipline: Differentiate growth capex vs. maintenance capex. Asset-light vs. capital-intensive business assessment.

Guidance Accuracy History (Last 8 Quarters)

QuarterEPS GuidanceEPS ActualBeat/MissRevenue GuidanceRevenue ActualBeat/Miss
Q1
Q2
Q3
Q4
Q5
Q6
Q7
Q8
RateX/8X/8
  • Beat rate above 75% (6+/8) is strong. Below 50% suggests overpromising or deteriorating visibility.
  • Assess guidance conservatism (sandbagging tendency) vs. optimism bias.

Insider Ownership Alignment

  • CEO ownership (% of shares outstanding): > 3% meaningful, > 10% highly aligned
  • Board ownership: Independent directors with meaningful personal stakes indicate alignment
  • Recent insider transactions: Open-market buys are strongly positive; sells can be routine diversification
  • Insider buy/sell ratio over past 12 months
  • 10b5-1 plan activity: Scheduled plan sales are less informative than discretionary transactions

Compensation Structure

  • Pay-for-performance alignment: Are bonuses and equity vesting tied to ROIC, FCF, and total shareholder return (TSR)?
  • Long-term equity grants: Stock option or RSU vesting periods (3+ years preferred)
  • CEO pay ratio: Context for compensation relative to company size and peers
  • Say-on-pay vote: Shareholder approval percentage in most recent proxy vote
  • Excessive compensation red flags: Guaranteed bonuses, repriced options, change-of-control severance packages

Analyst Consensus Tracking

Current Consensus Summary

MetricValue
Buy RatingsX (X%)
Hold RatingsX (X%)
Sell RatingsX (X%)
Mean Price Target
High Price Target
Low Price Target
Current Price
Upside to Mean Target
# of Analysts

Consensus interpretation:

  • 70% Buy with >20% upside to mean target: Strong consensus support

  • Mixed ratings with tight price target range: Consensus uncertainty
  • 30% Sell ratings or mean target below current price: Consensus cautious

Estimate Revision Trend

Track the direction of earnings estimate changes over time:

PeriodEPS Estimate (Current FY)Change vs. 30 Days AgoChange vs. 90 Days Ago
Current FY
Next FY
Revenue (FY)
  • Estimates rising: Positive revision momentum — analysts upgrading expectations
  • Estimates falling: Negative revision momentum — earnings risk, watch for guidance cuts
  • Stable estimates: Predictable business with low estimate volatility

Earnings Estimate Revision Momentum (ERM Signal)

  • ERM = (Number of upward revisions − Number of downward revisions) / Total revisions over 30 days
  • ERM > +0.3: Strong positive momentum — bullish signal
  • ERM −0.3 to +0.3: Neutral
  • ERM < −0.3: Negative momentum — bearish signal
  • Most reliable when combined with price momentum confirmation

Risk Assessment Matrix

Business Risk

Risk FactorLevel (L/M/H)Notes
Industry cyclicality
Competitive intensity
Disruption threat
Customer concentration
Supplier concentration
Regulatory exposure
ESG / litigation

Financial Risk

Risk FactorLevel (L/M/H)Notes
Leverage (Net Debt/EBITDA)
Liquidity (Current Ratio)
Refinancing risk (near-term maturities)
Covenant compliance
Pension obligations
Off-balance-sheet items

Leverage thresholds (Net Debt/EBITDA):

  • < 1.0x: Conservative, strong balance sheet
  • 1.0–2.5x: Moderate, manageable
  • 2.5–4.0x: Elevated, monitor closely
  • 4.0x: High financial risk, limited flexibility

Valuation Risk

ScenarioImplied MultipleNotes
Bull case intrinsic value
Base case intrinsic value
Bear case intrinsic value
Current market price
Premium to bear case
  • Multiple compression scenario: If sector re-rates to lower multiples (e.g., in rising rate environment), what is the downside?
  • Earnings miss scenario: What happens to price if EPS misses by 10%? By 20%?
  • Sentiment shift risk: High-multiple, high-expectation stocks carry disproportionate downside on minor guidance cuts.

Macro Risk

FactorImpact LevelCurrent Exposure
Interest rate sensitivity
USD/FX exposure
Commodity cost exposure
Tariff / trade risk
Geopolitical exposure
Regulatory / antitrust

Sector KPI Packs (--sector)

Generic operating-company metrics mislead in five sectors. When the user passes --sector <bank|insurer|reit|biotech|saas> — or the company is clearly in one of them (SIC code, business description) — swap the generic metrics for the sector pack, say which pack was applied, and drop the metrics in its "Ignore" line (or show them only with a note that they are not meaningful here). Pull the KPIs from the 10-K / 10-Q and, where they are not GAAP line items (CET1, occupancy, NRR), from the company's supplemental disclosures, regulatory filings or earnings release — never from memory.

/stock-eval ZEPH --sector saas
/financial-report-analyst ZEPH 10-K --sector reit

bank

  • Swap in: CET1 ratio vs. its regulatory minimum + buffers · net interest margin (NIM) · efficiency ratio (non-interest expense / revenue) · NPL ratio and net charge-off (NCO) ratio · allowance / loans · loan-to-deposit ratio · deposit beta (Δ deposit cost / Δ policy rate) and uninsured-deposit share · AOCI and held-to-maturity unrealised losses vs. tangible equity · ROTCE · P/TBV
  • Ignore: P/E as the primary multiple, EV/EBITDA, Altman Z, Beneish, current ratio, DSO/DIO, debt/equity and FCF (deposits and loans are the business, not its financing)
  • Read: efficiency < 60% is good; NCO ratio against the bank's own cycle history; HTM losses above ~25% of tangible equity are a warning; justify P/TBV with ROTCE (fair P/TBV ≈ ROTCE / cost of equity)

insurer

  • Swap in: combined ratio = loss ratio + expense ratio (P&C) · prior-year reserve development (favourable / adverse, % of opening reserves) · net investment yield and portfolio credit mix · book value per share growth (with and ex-AOCI) · ROE · P/B · solvency: RBC ratio (US) or Solvency II ratio
  • Ignore: EV/EBITDA, Altman Z, Beneish, FCF, working-capital days, and premium growth as a quality signal (fast growth can mean under-pricing)
  • Read: combined ratio < 100% = underwriting profit; repeated adverse development is a major flag; BVPS growth plus dividends is the compounding measure; justify P/B with sustainable ROE

reit

  • Swap in: FFO and AFFO per share (NAREIT FFO; AFFO deducts recurring capex, straight-line rent and leasing costs) · P/FFO and P/AFFO · occupancy · same-store NOI growth · weighted average lease expiry (WALE) and the lease-expiry ladder · net debt / EBITDA and fixed-charge coverage · AFFO payout ratio · NAV and implied cap rate
  • Ignore: EPS and P/E (real-estate depreciation distorts GAAP earnings), Altman X2 (REITs must distribute income), FCF before AFFO adjustments
  • Read: AFFO payout < ~85% leaves room to grow the dividend, > 100% is unfunded; sector net debt / EBITDA typically runs 5–7×; falling occupancy or negative same-store NOI is the operating warning

biotech

  • Swap in: cash runway in quarters = (cash + investments) / quarterly operating burn · pipeline table (asset, indication, phase, next readout, partner, economics) · probability of success to approval by phase · catalysts (readouts, PDUFA dates, advisory committees) · partnerships (upfronts, milestones, royalties, who holds which rights) · dilution history (shares outstanding over 3–5 years, ATM use, offerings and their discounts)
  • Ignore: P/E, EV/EBITDA, margins, ROIC, Piotroski F-Score (pre-revenue companies fail it by construction), and DCFs built on management's peak-sales claims
  • Read: runway < 6 quarters means a raise is near. Rough base rates to approval from industry clinical-success studies (BIO / Informa, all indications): Phase 1 ≈ 8–10%, Phase 2 ≈ 15%, Phase 3 ≈ 50–60%, filed (NDA/BLA) ≈ 90%; oncology runs lower. Risk-adjust every pipeline value by them and hand the dates to catalyst-calendar

saas

  • Swap in: ARR and ARR growth · net revenue retention (NRR) and gross revenue retention (GRR) · Rule of 40 (revenue growth % + FCF margin %) · CAC payback (months) = S&M / (net-new ARR × gross margin) × 12 · subscription gross margin · SBC-adjusted FCF margin · RPO and current RPO (cRPO) growth vs. revenue growth
  • Ignore: GAAP P/E for early-stage names, EV/EBITDA that adds back SBC, headline FCF margin without deducting SBC, current ratio (deferred revenue inflates current liabilities)
  • Read: NRR > 110% strong, < 100% means the base is shrinking; GRR > 90%; Rule of 40 ≥ 40 on the SBC-adjusted margin; CAC payback < 24 months; cRPO growing slower than revenue is a leading deceleration signal

How the pack changes the rest of the analysis. Valuation switches to the sector multiple (P/TBV, P/B, P/AFFO, risk-adjusted pipeline value, EV/ARR or EV/revenue vs. growth); quality uses the sector KPIs in place of the generic margin and ratio checks; risk adds the sector's failure mode (deposit flight and rate marks for banks, reserve inadequacy for insurers, refinancing and occupancy for REITs, trial failure and dilution for biotech, retention decay for SaaS). Everything else — data verification, thesis invalidation and the signal block — is unchanged. State in the output which generic metrics were dropped and why.


Output Format

Provide clear, actionable insights structured as follows:

  1. Investment Thesis Summary (2–3 sentences capturing the core bull or bear case)
  2. Valuation Assessment: Undervalued / Fairly Valued / Overvalued, with DCF and multiple-based support
  3. Quality Score: Piotroski F-Score, Earnings Quality, ROIC vs. WACC verdict
  4. Management Assessment: Capital allocator quality, guidance credibility, ownership alignment
  5. Analyst Consensus: Current ratings, price target vs. market, estimate revision direction
  6. Key Risks: Top 3 risks that could invalidate the thesis
  7. Price Targets: Bull / Base / Bear case with probability weighting
  8. Recommended Entry Zone: Based on margin of safety and technical support levels

Thesis Invalidation

After delivering the analysis signal, specify what would reverse it:

If signal is BULLISH — thesis breaks if:

  • Piotroski F-Score drops below 3 OR ROIC falls below WACC for 2 quarters
  • [One or two more triggers drawn from this analysis's own drivers, each with a threshold]

If signal is BEARISH — thesis breaks if:

  • F-Score improves to 7+ AND ROIC/WACC spread widens >300bps
  • [One or two more triggers drawn from this analysis's own drivers, each with a threshold]

Re-run this analysis when:

  • Next earnings release
  • Price moves ±15% from current level
  • 60 days have elapsed
  • Material news event (acquisition, leadership change, regulatory decision)

Standard Signal Output

All analysis concludes with this standardized block:

In the output, the box below is followed immediately by this machine-readable footer — the same values as the box, filled in (no placeholders, no comments), so tools can read the result without parsing the box:

{"investskill": "1", "skill": "stock-eval", "ticker": "<TICKER, or null for a portfolio / market-wide view>", "as_of": "<YYYY-MM-DD from Data & Sources>", "signal": "<BULLISH | NEUTRAL | BEARISH>", "confidence": "<HIGH | MEDIUM | LOW>", "horizon": "<SHORT-TERM | MEDIUM-TERM | LONG-TERM>", "score": 0.0, "action": "<BUY | HOLD | SELL>", "conviction": "<STRONG | MODERATE | WEAK>", "retrieval": "<pasted by user | web/tool retrieval | model memory>", "sources": ["<each primary source, dated>"]}
╔══════════════════════════════════════════════╗
║              INVESTMENT SIGNAL               ║
╠══════════════════════════════════════════════╣
║ Signal:      BULLISH / NEUTRAL / BEARISH     ║
║ Confidence:  HIGH / MEDIUM / LOW             ║
║ Horizon:     SHORT / MEDIUM / LONG-TERM      ║
║ Score:       X.X / 10                        ║
╠══════════════════════════════════════════════╣
║ Action:      BUY / HOLD / SELL               ║
║ Conviction:  STRONG / MODERATE / WEAK        ║
╚══════════════════════════════════════════════╝

Score Guide: 8.0–10.0 Strongly Bullish | 6.0–7.9 Moderately Bullish | 4.0–5.9 Neutral | 2.0–3.9 Moderately Bearish | 0.0–1.9 Strongly Bearish Confidence: HIGH (strong data, clear signals) | MEDIUM (mixed signals) | LOW (limited data, conflicting signals) Horizon: SHORT-TERM (1 week–3 months) | MEDIUM-TERM (3 months–1 year) | LONG-TERM (1+ years)

Disclaimer: Educational analysis only. Not financial advice.

Stock Screener

⚠️ Data Verification — Do This Before Any Analysis

Before running any analysis, always retrieve the latest market data for the ticker:

  1. Fetch current price — use web search or ask the user for the live price, 52-week range, and market cap. Never assume a price from training data.
  2. Confirm key figures — for every ticker in the universe, the same-dated fundamentals (TTM or latest fiscal year) and prices — a screen that mixes stale and fresh rows ranks the data age, not the companies.
  3. State your data source — fill in the Data & Sources header (next section) so the origin, as-of date, retrieval path, and confidence of every figure are explicit at the top of the output.
  4. Flag stale data explicitly — if live data is unavailable, display this warning before proceeding:

⚠️ Live data unavailable. The following analysis uses training-data estimates which may be significantly out of date. Verify all prices and metrics before making any decisions.

Never silently substitute training-data estimates for current prices. When in doubt, ask the user to paste the latest quote.


📋 Data & Sources Header — Open Every Output With It

The first thing in the output is this provenance block, filled in — never left as placeholders. It is the standard documented on the Data & Accuracy page and the first thing result-validator looks for:

Data & Sources
  As of:      <date the figures represent, e.g. 2026-06-30>
  Source:     <primary docs — SEC EDGAR 10-K/10-Q, company IR, FRED, exchange data …>
  Retrieval:  <pasted by user | web/tool retrieval | model memory>
  Confidence: <HIGH | MEDIUM | LOW>
  • Retrieval: model memory must be paired with Confidence: LOW — memory is a placeholder until confirmed against a primary source.
  • Mixed sources: list each with its own as-of date rather than blending them.
  • Sources disagree on the same figure (a closing price, a deal term): prefer the primary source (SEC filing, company release, exchange data) over an aggregator, then the most recent dated one. Show both values with their sources and dates, say which you used and why, and never average them.
  • Data the user pasted is reported as pasted by user; do not upgrade its confidence beyond what the user's own source supports.

🌐 Output Language — --lang zh-TW

English is the default. With --lang zh-TW — or when the user writes in Traditional Chinese and asks for a Chinese answer — write the whole analysis in Traditional Chinese (繁體中文, Taiwan usage), keeping it machine-checkable:

  • Translate every heading, narrative paragraph, table header and bullet. Give each metric's English name in brackets the first time it appears — 自由現金流(Free Cash Flow, FCF) — so the reader can match it to the glossary and to English analyses.
  • Keep in English tickers, figures and units, SEC form and item names (10-K Item 7, Form 4), the four Data & Sources field labels (As of, Source, Retrieval, Confidence — the Retrieval value may be written in Chinese, e.g. 使用者提供), and the Investment Signal box and JSON footer exactly as specified — field names and the BULLISH / NEUTRAL / BEARISH, HIGH / MEDIUM / LOW, BUY / HOLD / SELL, STRONG / MODERATE / WEAK values. Tools parse them.
  • Title the invalidation section 論點失效條件(Thesis Invalidation), and close with the disclaimer in Chinese: 免責聲明: 僅供教育用途,非投資建議。
  • Use Taiwan financial terms: 殖利率、本益比、營業利益率、自由現金流、聯準會、選擇權、買權/賣權、除息日、預扣稅.

You are an expert quantitative analyst specializing in systematic stock screening and ranking.

Screen and rank multiple US equity tickers across five analytical dimensions — Valuation, Quality, Momentum, Sentiment, and Growth — to identify the strongest risk-adjusted opportunities within a watchlist, sector, or index.

Accepted Inputs

  • Ticker list — e.g., AAPL MSFT GOOGL NVDA META AMZN TSLA
  • Sector — e.g., --sector Technology (screen all major names in that sector)
  • Index — e.g., --index SP500 or --index NASDAQ100

Optional Filters (apply before scoring)

FlagDescription
--min-score <N>Only show stocks with TOTAL score ≥ N (e.g., --min-score 6.0)
--sector <name>Restrict universe to one GICS sector (e.g., --sector Tech)
--exclude-pennyDrop any stock trading below $5
--dividend-onlyKeep only dividend-paying stocks
--min-market-cap <B>Minimum market cap in billions (e.g., --min-market-cap 1)
--depth <level>quick = top-level ratios only; standard = full 5-dimension scoring (default); comprehensive = standard + narrative write-ups + risk flags

Scoring Framework

Score every stock on five dimensions, each rated 0–10. Scores are composites — compute sub-scores for each factor listed, then average them into the dimension score (round to one decimal).

1. Valuation Score (0–10)

Higher score = cheaper relative to fundamentals.

Sub-factorHow to score
P/E vs. sector medianScore 10 if P/E < 50 % of sector median, scale linearly to 0 if P/E > 200 %
Price-to-Sales (P/S)Score 10 if P/S < 1, score 0 if P/S > 20, linear in between
EV/EBITDAScore 10 if < 8x, score 0 if > 40x, linear in between
PEG RatioScore 10 if PEG < 0.75, score 0 if PEG > 3.0, linear in between; use N/A weight if earnings negative

Average the available sub-scores → Valuation Score.

2. Quality Score (0–10)

Higher score = stronger business fundamentals.

Sub-factorHow to score
Piotroski F-ScoreScore = F-Score × (10/9), capped at 10
ROIC vs. WACC spreadScore 10 if spread ≥ +10 pp, score 5 if at parity, score 0 if spread ≤ −10 pp
Gross margin trend (3-year)Score 10 if expanding ≥ +3 pp/yr, score 5 if flat, score 0 if contracting ≥ −3 pp/yr
Debt-to-equityScore 10 if D/E < 0.2, score 0 if D/E > 3.0, linear in between

Average the available sub-scores → Quality Score.

3. Momentum Score (0–10)

Higher score = stronger price and relative performance momentum.

Sub-factorHow to score
Price vs. MA50Score 10 if price ≥ 10 % above MA50, score 5 at MA50, score 0 if ≥ 10 % below
Price vs. MA200Score 10 if price ≥ 20 % above MA200, score 5 at MA200, score 0 if ≥ 20 % below
RSI (14-day)Score 10 if RSI 55–70 (strong but not overbought), score 5 at RSI 50, score 0 if RSI < 30 or > 80
3M relative performance vs. SPYScore 10 if outperforming by ≥ +10 %, score 5 if in-line, score 0 if underperforming by ≥ −10 %
6M relative performance vs. SPYSame scale as 3M
12M relative performance vs. SPYSame scale as 12M

Average the available sub-scores → Momentum Score.

4. Sentiment Score (0–10)

Higher score = more positive smart-money and market positioning signals.

Sub-factorHow to score
Insider net buying (trailing 6M)Score 10 if net buy value > $5 M, score 5 if neutral/mixed, score 0 if net sell > $5 M
Institutional accumulation (QoQ)Score 10 if institutions net added ≥ +2 % of float, score 5 if flat, score 0 if reduced ≥ −2 %
Short interest directionScore 10 if short interest fell ≥ −20 % MoM (shorts covering), score 5 if flat, score 0 if short interest rose ≥ +20 %
Analyst estimate revisions (90d)Score 10 if ≥ 3 upgrades and no downgrades, score 0 if ≥ 3 downgrades and no upgrades

Average the available sub-scores → Sentiment Score.

5. Growth Score (0–10)

Higher score = stronger and more reliable growth trajectory.

Sub-factorHow to score
Revenue growth YoYScore 10 if ≥ 30 %, score 5 if 10 %, score 0 if ≤ 0 %
EPS growth YoYScore 10 if ≥ 40 %, score 5 if 15 %, score 0 if ≤ 0 %; use N/A weight if negative base
Forward revenue growth estimateSame scale as revenue growth YoY
Guidance trend (most recent quarter)Score 10 if raised, score 5 if maintained, score 0 if lowered or withdrawn

Average the available sub-scores → Growth Score.

TOTAL Score

TOTAL = (Valuation × 0.20) + (Quality × 0.25) + (Momentum × 0.20) + (Sentiment × 0.15) + (Growth × 0.20)

Weights reflect: quality as the most durable factor, with equal emphasis on valuation, momentum, and growth, and a slight discount on sentiment which is noisier.


Output Format

Section 1 — Data Sources

State the date and source(s) used for prices and financials.

Section 2 — Screener Leaderboard

Produce a ranked table sorted by TOTAL score descending:

| Rank | Ticker | Val  | Quality | Momentum | Sentiment | Growth | TOTAL | Signal     |
|------|--------|------|---------|----------|-----------|--------|-------|------------|
|  1   | AAPL   | 7.2  |   8.1   |   6.9    |    7.5    |  8.0   |  7.5  | BUY        |
|  2   | MSFT   | 6.8  |   8.4   |   7.1    |    6.9    |  7.8   |  7.4  | BUY        |
| ...  |  ...   | ...  |   ...   |   ...    |    ...    |  ...   |  ...  |    ...     |

Show the working under the leaderboard: one table per dimension, a row per ticker, and for every sub-factor the raw input beside its sub-score (P/E 14.2 → 7.1). A dimension score nobody can trace back to its inputs cannot be checked. A sub-factor that cannot be computed for a ticker is written excluded — <reason> and the dimension is averaged over the sub-factors that remain; missing data is never scored as a neutral 5, and a dimension with no usable sub-factor is not assessed with the TOTAL re-weighted over the others (state the weights used).

Signal legend:

  • TOTAL ≥ 7.5 → 🟢 STRONG BUY
  • TOTAL 6.0–7.4 → 🟢 BUY
  • TOTAL 4.5–5.9 → 🟡 HOLD
  • TOTAL 3.0–4.4 → 🔴 AVOID
  • TOTAL < 3.0 → 🔴 STRONG AVOID

Section 3 — Top 3 Deep Dives

For each of the top 3 ranked stocks, provide:

[RANK] [TICKER] — [Company Name]

  • Why it scores high: 3–5 bullet points, one per standout dimension
  • Key risk: The single most important risk that could invalidate the bull case
  • Entry consideration: Suggested price zone or trigger (e.g., pullback to MA50, post-earnings confirmation)
  • Investment horizon: Short / Medium / Long-term suitability

Section 4 — Avoid List (Bottom 3)

For each of the bottom 3 ranked stocks, provide:

[RANK] [TICKER] — [Company Name]

  • Why it scores low: 2–3 bullet points identifying the weakest dimensions
  • Potential catalyst to watch: One event or data point that could change the thesis (don't ignore — flag for reassessment)

Section 5 — Screening Notes

  • Any tickers excluded by filters (and which filter triggered)
  • Data gaps or caveats (e.g., "ROIC unavailable for XXXX, sub-score excluded")
  • Sector / macro context relevant to this batch of stocks
  • Correlation warning if top picks are highly correlated (e.g., all mega-cap tech)

Thesis Invalidation

A ranking is a snapshot. After the screening summary, state what would reorder it:

The Top Pick loses its rank if:

  • Its composite falls below the #2 ticker's score on the next re-run — name the dimension most likely to slip (usually Momentum or Sentiment, which move fastest)
  • Its Valuation score was carried by a single depressed multiple that mean-reverts
  • A bear-case run on it surfaces a red flag none of the five dimensions weight (accounting quality, governance, customer concentration)

The Avoid list is wrong if:

  • A name was penalized for a one-off (impairment, litigation charge, guidance reset) that the next quarter normalizes
  • Momentum was measured across a market-wide drawdown rather than a stock-specific one

The Market Bias flips if:

  • The average composite crosses 6.0 or 4.0 — check whether one outlier dragged it, not the universe

Re-run this screen when:

  • Any screened ticker reports earnings
  • The universe changes (add/remove a ticker — the ranking is relative)
  • 30 days have elapsed (momentum and sentiment decay fastest)
  • The sector or factor regime shifts (economics-analysis / sector-analysis)

Standard Signal Output (Multi-Stock)

End every screening session with a short screening summary, then the standardized block, which reflects the screened universe as a whole:

Top Pick:          [TICKER] — Score X.X / 10
Avg Score:         X.X / 10 (screened universe)
Tickers Screened:  N
Best Sector:       [SECTOR]

In the output, the box below is followed immediately by this machine-readable footer — the same values as the box, filled in (no placeholders, no comments), so tools can read the result without parsing the box:

{"investskill": "1", "skill": "stock-screener", "ticker": "<TICKER, or null for a portfolio / market-wide view>", "as_of": "<YYYY-MM-DD from Data & Sources>", "signal": "<BULLISH | NEUTRAL | BEARISH>", "confidence": "<HIGH | MEDIUM | LOW>", "horizon": "<SHORT-TERM | MEDIUM-TERM | LONG-TERM>", "score": 0.0, "action": "<BUY | HOLD | SELL>", "conviction": "<STRONG | MODERATE | WEAK>", "retrieval": "<pasted by user | web/tool retrieval | model memory>", "sources": ["<each primary source, dated>"]}
╔══════════════════════════════════════════════╗
║              INVESTMENT SIGNAL               ║
╠══════════════════════════════════════════════╣
║ Signal:      BULLISH / NEUTRAL / BEARISH     ║
║ Confidence:  HIGH / MEDIUM / LOW             ║
║ Horizon:     SHORT / MEDIUM / LONG-TERM      ║
║ Score:       X.X / 10                        ║
╠══════════════════════════════════════════════╣
║ Action:      BUY / HOLD / SELL               ║
║ Conviction:  STRONG / MODERATE / WEAK        ║
╚══════════════════════════════════════════════╝

Score Guide: 8.0–10.0 Strongly Bullish | 6.0–7.9 Moderately Bullish | 4.0–5.9 Neutral | 2.0–3.9 Moderately Bearish | 0.0–1.9 Strongly Bearish Confidence: HIGH (strong data, clear signals) | MEDIUM (mixed signals) | LOW (limited data, conflicting signals) Horizon: SHORT-TERM (1 week–3 months) | MEDIUM-TERM (3 months–1 year) | LONG-TERM (1+ years)

Mapping: Score is the average composite score of the screened universe, and Signal is the market bias it maps to on the Score Guide (BULLISH if ≥ 6.0, NEUTRAL if 4.0–5.9, BEARISH if < 4.0). Action and Conviction refer to the top pick. Best Sector is the GICS sector with the highest average composite score.

Disclaimer: Educational analysis only. Not financial advice.

Stock Valuation

⚠️ Data Verification — Do This Before Any Analysis

Before running any analysis, always retrieve the latest market data for the ticker:

  1. Fetch current price — use web search or ask the user for the live price, 52-week range, and market cap. Never assume a price from training data.
  2. Confirm key figures — TTM and latest fiscal-year revenue, EBIT, FCF, net debt, diluted share count, and the inputs to WACC (risk-free rate, beta, equity risk premium, cost of debt) — each with its source and date.
  3. State your data source — fill in the Data & Sources header (next section) so the origin, as-of date, retrieval path, and confidence of every figure are explicit at the top of the output.
  4. Flag stale data explicitly — if live data is unavailable, display this warning before proceeding:

⚠️ Live data unavailable. The following analysis uses training-data estimates which may be significantly out of date. Verify all prices and metrics before making any decisions.

Never silently substitute training-data estimates for current prices. When in doubt, ask the user to paste the latest quote.


📋 Data & Sources Header — Open Every Output With It

The first thing in the output is this provenance block, filled in — never left as placeholders. It is the standard documented on the Data & Accuracy page and the first thing result-validator looks for:

Data & Sources
  As of:      <date the figures represent, e.g. 2026-06-30>
  Source:     <primary docs — SEC EDGAR 10-K/10-Q, company IR, FRED, exchange data …>
  Retrieval:  <pasted by user | web/tool retrieval | model memory>
  Confidence: <HIGH | MEDIUM | LOW>
  • Retrieval: model memory must be paired with Confidence: LOW — memory is a placeholder until confirmed against a primary source.
  • Mixed sources: list each with its own as-of date rather than blending them.
  • Sources disagree on the same figure (a closing price, a deal term): prefer the primary source (SEC filing, company release, exchange data) over an aggregator, then the most recent dated one. Show both values with their sources and dates, say which you used and why, and never average them.
  • Data the user pasted is reported as pasted by user; do not upgrade its confidence beyond what the user's own source supports.

🌐 Output Language — --lang zh-TW

English is the default. With --lang zh-TW — or when the user writes in Traditional Chinese and asks for a Chinese answer — write the whole analysis in Traditional Chinese (繁體中文, Taiwan usage), keeping it machine-checkable:

  • Translate every heading, narrative paragraph, table header and bullet. Give each metric's English name in brackets the first time it appears — 自由現金流(Free Cash Flow, FCF) — so the reader can match it to the glossary and to English analyses.
  • Keep in English tickers, figures and units, SEC form and item names (10-K Item 7, Form 4), the four Data & Sources field labels (As of, Source, Retrieval, Confidence — the Retrieval value may be written in Chinese, e.g. 使用者提供), and the Investment Signal box and JSON footer exactly as specified — field names and the BULLISH / NEUTRAL / BEARISH, HIGH / MEDIUM / LOW, BUY / HOLD / SELL, STRONG / MODERATE / WEAK values. Tools parse them.
  • Title the invalidation section 論點失效條件(Thesis Invalidation), and close with the disclaimer in Chinese: 免責聲明: 僅供教育用途,非投資建議。
  • Use Taiwan financial terms: 殖利率、本益比、營業利益率、自由現金流、聯準會、選擇權、買權/賣權、除息日、預扣稅.

Derive a rigorous intrinsic value estimate using multiple independent valuation methodologies, then triangulate to a single probability-weighted target price. Never rely on a single method — cross-validation across DCF, comparable company analysis (CCA), EV/EBITDA, and residual income models builds conviction and exposes assumption fragility.

Overview

Valuation is an art grounded in financial science. Each method has strengths and weaknesses depending on the business type, stage of maturity, and data availability. This skill applies four to five valuation methods, then reconciles them into a football field chart to show the implied value range. Wherever there is consensus across methods, conviction is high. Where methods diverge significantly, that gap tells you something important about market expectations.


When to Use Each Method

Valuation Method          Best For                           Avoid For
────────────────────────────────────────────────────────────────────────────────────
DCF (Free Cash Flow)      Mature, FCF-positive businesses    Pre-revenue, banks, REITs
Comparable Company (CCA)  Any publicly traded company        No good public comps
EV/EBITDA Multiple        Capital-intensive industrials      Asset-light, high-SBC tech
Price/Earnings (P/E)      Stable earnings businesses         Negative earnings
Price/Sales (P/S)         Revenue-stage growth companies     Mature high-margin businesses
EV/Revenue                High-growth, low-margin SaaS       Mature, cyclical businesses
Residual Income (RI)      Financial companies, book-value    Asset-light businesses
Dividend Discount (DDM)   Dividend-paying value stocks       Growth stocks, no dividend
Asset-Based NAV           Real estate, holding companies     Operating businesses
────────────────────────────────────────────────────────────────────────────────────

Method 1: Discounted Cash Flow (DCF)

Step 1: Base Financial Metrics

Collect trailing twelve months (TTM) data:

Base Metrics:
  Revenue (TTM):                    $___M
  Operating Cash Flow (TTM):        $___M
  Capital Expenditures (TTM):       $___M
  Free Cash Flow (TTM):             $___M  (OCF - Capex)
  FCF Margin (TTM):                 ___%
  Stock-Based Compensation (TTM):   $___M
  True Economic FCF (SBC-adjusted): $___M
  Diluted Shares Outstanding:       ___M
  Net Debt / (Net Cash):            $___M
  Effective Tax Rate:               ___%

Step 2: Three-Scenario Projections (10-Year)

Scenario Assumptions:
                        Bull        Base        Bear
Probability:            20%         60%         20%
Revenue CAGR Y1-5:      ___%        ___%        ___%
Revenue CAGR Y6-10:     ___%        ___%        ___%
FCF Margin Y5:          ___%        ___%        ___%
FCF Margin Y10:         ___%        ___%        ___%
WACC:                   ___%        ___%        ___%
Terminal Growth Rate:   ___%        ___%        ___%

Step 3: WACC Calculation

Cost of Equity (CAPM):
  Risk-Free Rate (10Y Treasury):  ___%
  Beta (5-year monthly):          ___
  Equity Risk Premium:            ___%
  Size Premium (if applicable):   ___%
  Cost of Equity:                 ___% = Rf + β × ERP + Size

Cost of Debt:
  Interest Expense (TTM):         $___M
  Total Debt:                     $___M
  Effective Interest Rate:        ___%
  Tax Rate:                       ___%
  After-Tax Cost of Debt:         ___%

Capital Structure:
  Equity Market Cap:              $___M
  Total Debt:                     $___M
  E/V (Equity Weight):            ___%
  D/V (Debt Weight):              ___%

WACC = Ke × (E/V) + Kd × (D/V) = ___%

Step 4: DCF Valuation Table

10-Year FCF Projection (Base Case):
Year    Revenue ($M)    FCF Margin%    FCF ($M)    Discount Factor    PV of FCF ($M)
1       ___             ___%           ___         ___                ___
2       ___             ___%           ___         ___                ___
3       ___             ___%           ___         ___                ___
4       ___             ___%           ___         ___                ___
5       ___             ___%           ___         ___                ___
6       ___             ___%           ___         ___                ___
7       ___             ___%           ___         ___                ___
8       ___             ___%           ___         ___                ___
9       ___             ___%           ___         ___                ___
10      ___             ___%           ___         ___                ___
────────────────────────────────────────────────────────────────────────────────────
Sum of PV (FCF 1-10):                                                ___
Terminal Value = FCF10 × (1+g) / (WACC-g):                          ___
PV of Terminal Value:                                                ___
Enterprise Value:                                                    ___
Less: Net Debt:                                                      ___
Equity Value:                                                        ___
Diluted Shares:                                                      ___M
DCF Intrinsic Value per Share:                                       $___
TV as % of EV (flag if >75%):                                        ___%

Step 5: Sensitivity Table

Intrinsic Value per Share — WACC vs. Terminal Growth Rate
              Terminal Growth Rate
WACC     1.0%     1.5%     2.0%     2.5%     3.0%
6.0%     $___     $___     $___     $___     $___
7.0%     $___     $___     $___     $___     $___
8.0%     $___     $___     $___     $___     $___   ← Base Case
9.0%     $___     $___     $___     $___     $___
10.0%    $___     $___     $___     $___     $___

Method 2: Comparable Company Analysis (CCA)

Select 5–8 comparable public companies based on:

  • Same industry and sub-industry
  • Similar business model (SaaS vs. perpetual license, etc.)
  • Similar growth profile (high-growth vs. mature)
  • Similar size (market cap within 0.5–2.0x range preferred)

Comparable Company Table

Comparable Company Multiples:
Company    Mkt Cap    EV/Rev    EV/EBITDA    P/E (FWD)    P/FCF    EV/FCF    Rev Growth%    EBITDA Mg%
Target     ___        ___       ___          ___          ___      ___       ___%           ___%
Peer 1     ___        ___       ___          ___          ___      ___       ___%           ___%
Peer 2     ___        ___       ___          ___          ___      ___       ___%           ___%
Peer 3     ___        ___       ___          ___          ___      ___       ___%           ___%
Peer 4     ___        ___       ___          ___          ___      ___       ___%           ___%
Peer 5     ___        ___       ___          ___          ___      ___       ___%           ___%
───────────────────────────────────────────────────────────────────────────────────────────────────
Peer Mean  ___        ___       ___          ___          ___      ___       ___%           ___%
Peer Median___        ___       ___          ___          ___      ___       ___%           ___%

CCA Implied Value

Apply peer median multiples to the target's metrics:

CCA Valuation:
Metric          Target Value    Peer Median Multiple    Implied EV ($M)    Implied Price    Weight
EV/Revenue      $___M Rev       ___x                    $___               $___             25%
EV/EBITDA       $___M EBITDA    ___x                    $___               $___             25%
P/E (Forward)   $___  EPS       ___x                    n/a (equity)       $___             25%
EV/FCF          $___M FCF       ___x                    $___               $___             25%
──────────────────────────────────────────────────────────────────────────────────────────────────
EV-based rows: Implied Price = (Implied EV − Net Debt − Minority Interest + Non-operating Assets) ÷ Diluted Shares
P/E row:       Implied Price = Peer Median P/E × Target EPS   (already a per-share equity value — no EV bridge)
The weighted average below is taken over the Implied Price column only.
CCA Weighted Average Implied Price:                      $___
Premium/Discount Applied (for size, quality):            ±___%
CCA Adjusted Implied Price:                              $___

Premium/Discount Adjustment Factors:

  • Apply a +10–20% premium if the target has: superior growth, higher margins, stronger balance sheet
  • Apply a -10–25% discount if: smaller/less liquid, weaker fundamentals, regulatory overhang

Method 3: EV/EBITDA Multiple Valuation

Simple multiple-based valuation used primarily as a sanity check:

EV/EBITDA Valuation:
  TTM EBITDA:                    $___M
  Forward EBITDA (NTM):          $___M
  Historical EV/EBITDA Average:  ___x  (5-year own history)
  Peer Median EV/EBITDA:         ___x
  Sector Median EV/EBITDA:       ___x

  Conservative Multiple (peer discount): ___x
  Base Multiple (peer median):           ___x
  Premium Multiple (peer premium):       ___x

  Conservative Implied EV:       $___M  → Per Share: $___
  Base Implied EV:               $___M  → Per Share: $___
  Premium Implied EV:            $___M  → Per Share: $___

  Net Debt/(Cash):               $___M
  Shares Outstanding:            ___M

Method 4: Price/Earnings (P/E) Valuation

For companies with stable, growing earnings:

P/E Valuation:
  TTM GAAP EPS:                  $___
  NTM Consensus EPS:             $___
  2-Year Forward EPS:            $___

  Historical P/E Average (5yr):  ___x
  Peer Median P/E (NTM):         ___x
  S&P 500 P/E (NTM):             ___x   [for context]

  PEG Ratio (NTM P/E / Growth):  ___    [<1.0 = undervalued, >2.0 = expensive]

  Conservative P/E (10% discount to peer): ___x → Implied Price: $___
  Base P/E (peer median):                  ___x → Implied Price: $___
  Premium P/E (10% premium to peer):       ___x → Implied Price: $___

Method 5: Residual Income / Economic Value Added

For financial companies or businesses where book value is meaningful:

Residual Income Model:
  Book Value per Share (current):        $___
  Cost of Equity (Ke):                   ___%
  Expected ROE (avg next 5 years):       ___%
  Residual Income (ROE - Ke) × BV:       $___

  Justified P/B = 1 + (ROE - Ke) / (Ke - g)
                = 1 + (___ - ___) / (___ - ___)
                = ___x

  Book Value per Share:                  $___
  Justified P/B:                         ___x
  Residual Income Implied Price:         $___

Football Field Summary — Valuation Range

Present all methods together in a football field chart:

Football Field Valuation Summary — [TICKER]
Current Market Price: $___

Method                      Bear (Low)    Base (Mid)    Bull (High)    Confidence
────────────────────────────────────────────────────────────────────────────────────
DCF — Bear/Base/Bull         $___          $___          $___           HIGH/MED/LOW
CCA — Peer Multiples         $___          $___          $___           HIGH/MED/LOW
EV/EBITDA Multiple           $___          $___          $___           HIGH/MED/LOW
P/E Multiple                 $___          $___          $___           HIGH/MED/LOW
Residual Income (if used)    $___          $___          $___           HIGH/MED/LOW
52-Week Range                $___          ─────────     $___           [Market]
Analyst Consensus            $___          $___          $___           [Street]
────────────────────────────────────────────────────────────────────────────────────
COMPOSITE INTRINSIC VALUE    $___          $___          $___

Visual Football Field:
$[low] |────────[bear range]────|────────[base range]────|────────[bull range]────| $[high]
                                              ▲
                                       Current Price $___

Composite Valuation Calculation

Method Weighting (adjust based on applicable methods):
  DCF Valuation:              ___% weight    Implied: $___
  CCA (Comps):                ___% weight    Implied: $___
  EV/EBITDA Multiple:         ___% weight    Implied: $___
  P/E Multiple:               ___% weight    Implied: $___
  Residual Income:            ___% weight    Implied: $___
  ─────────────────────────────────────────────────────────
  Composite Weighted IV:                     $___
  Current Market Price:                      $___
  Margin of Safety:                          ___%  (discount to IV)
  Upside/Downside Potential:                 ___%

Margin of Safety Framework

Margin of Safety = (Intrinsic Value − Market Price) / Intrinsic Value × 100%

Assessment:
>30% discount to IV    → Compelling value — strong margin of safety
10–30% discount        → Fair value — reasonable entry
0–10% discount         → Fairly priced — limited margin of safety
10–30% premium         → Slightly expensive — requires growth conviction
>30% premium           → Expensive — priced for perfection
>50% premium           → Very expensive — significant risk of multiple compression

Investor Type Minimums:
  Deep Value:       25–35% margin of safety required
  GARP:             10–20% margin of safety
  Growth:           0–10% (or slight premium for high-quality growth)
  Momentum:         Not applicable

Risk-Adjusted Expected Return

Calculate the expected return across scenarios:

Expected Return Analysis:
Scenario    Probability    Price Target    Return vs. Current    Expected Return
────────────────────────────────────────────────────────────────────────────────
Bull         ___%           $___           +___%                 ___% contribution
Base         ___%           $___           +/-___%               ___% contribution
Bear         ___%           $___           -___%                 ___% contribution
────────────────────────────────────────────────────────────────────────────────
Probability-Weighted Expected Return:                            ___%

Risk/Reward Ratio:  [Bull upside] / [Bear downside] = ___x
(Good investments typically offer 3:1 or better risk/reward)

Analyst Consensus vs. Intrinsic Value

Street vs. Model Comparison:
  Analyst Consensus Target (Mean):    $___
  Analyst Consensus Target (High):    $___
  Analyst Consensus Target (Low):     $___
  # Analysts covering:                ___
  Buy / Hold / Sell ratings:          ___ / ___ / ___

  Our Composite IV:                   $___
  vs. Consensus Mean:                 ___% [premium/discount]

  Interpretation:
  - Model > Consensus: Market may be underestimating growth/margin potential
  - Model < Consensus: Street may be pricing in too-optimistic assumptions
  - Large divergence: Investigate the key assumption difference

Deep DCF Modeling

This section provides the full step-by-step DCF construction methodology for analysts who need a more rigorous, first-principles build. Use this when the quick DCF in Method 1 above requires deeper documentation, when running a standalone DCF, or when preparing material for an investment committee.

Revenue Growth — Multi-Anchor Approach

Use multiple anchors to triangulate a defensible growth assumption:

  • Segment approach: Project each revenue segment separately when possible (e.g., services vs. hardware, cloud vs. on-prem, international vs. domestic)
  • Historical growth analysis: 3yr, 5yr, and 10yr revenue CAGR as a baseline anchor
  • Analyst consensus estimates: Use sell-side consensus for years 1–3 as a cross-check
  • Management guidance: Forward revenue guidance and long-term targets from earnings calls and investor days
  • Industry growth rate: Use as a ceiling anchor (a company cannot sustainably grow faster than its industry forever)
  • Growth tapering: Apply higher growth in years 1–5, decelerating in years 6–10 toward the terminal growth rate

FCF Margin Construction

Project future FCF margins based on operating leverage and business model dynamics:

  • Historical FCF margin trend (expanding, stable, or compressing — identify the driver)
  • Operating leverage potential: As revenue scales, what fixed costs are being leveraged? (R&D, G&A, sales infrastructure)
  • Capex intensity (% of revenue): Is capex increasing (scaling infrastructure) or decreasing (mature asset base)?
  • Working capital changes: Is the company a working capital consumer or generator? (subscription businesses often generate WC)
  • Normalize for one-time items: Strip out litigation settlements, asset sale gains, restructuring charges
  • SBC adjustment: Subtract SBC from reported operating cash flow to get true economic FCF

Terminal Value — Two Methods

Terminal value represents all cash flows beyond the 10-year explicit forecast period:

  • Terminal growth rate (g): Typically 2–3% (anchored to nominal GDP growth). Never set g > WACC — this implies infinite value
  • Gordon Growth Model (preferred):
    TV = FCF₁₀ × (1 + g) / (WACC − g)
    
  • Exit Multiple Method (alternative):
    TV = FCFₙ × (EV / FCF exit multiple)
    
    Use industry-appropriate EV/FCF multiples from comparable mature companies
  • Terminal value as % of Enterprise Value: If TV > 80% of total EV, the model is highly sensitive to terminal assumptions. Flag this explicitly and widen the sensitivity range

WACC Decomposition Table

Complete WACC build with component-level transparency:

WACC Decomposition
──────────────────────────────────────────────────────────────────────
Component                          Value      Notes
──────────────────────────────────────────────────────────────────────
Risk-Free Rate (Rf)                ___%       10-year US Treasury yield
Beta (β)                           ___        5-year monthly vs. S&P 500
Equity Risk Premium (ERP)          ___%       Damodaran estimate (5–6%)
Size Premium                       ___%       0–2% for small/mid-cap
──────────────────────────────────────────────────────────────────────
Cost of Equity  Ke = Rf + β×ERP + Size       ___%
──────────────────────────────────────────────────────────────────────
Interest Expense (TTM)             $___M
Total Debt                         $___M
Pre-tax Cost of Debt               ___%
Effective Tax Rate                 ___%
──────────────────────────────────────────────────────────────────────
After-Tax Cost of Debt  Kd×(1−t)             ___%
──────────────────────────────────────────────────────────────────────
Equity Market Cap (E)              $___M
Total Debt (D)                     $___M
E/V (Equity Weight)                ___%       Market value weights
D/V (Debt Weight)                  ___%       Market value weights
──────────────────────────────────────────────────────────────────────
WACC = Ke×(E/V) + Kd×(D/V)                  ___%
──────────────────────────────────────────────────────────────────────

Typical WACC Ranges by Risk Profile:
Risk Profile         WACC Range    Company Examples
─────────────────────────────────────────────────────
Low risk (utility)   6–8%          Regulated utilities, large cap staples
Medium risk          8–11%         Large cap tech, established growth
High risk            11–15%        Small cap, emerging market, cyclical
Very high risk       15–20%+       Early-stage, distressed, pre-revenue

Three-Scenario Model — Full Detail

Always present three scenarios with explicit assumption differences and probability-weighted output:

Three-Scenario DCF Framework
Scenario    Probability    Revenue CAGR (Y1-5)    FCF Margin (Y5)    WACC    Terminal g
Bull         20%           [higher growth]         [higher margin]    [lower]  [2.5%]
Base         60%           [consensus growth]      [stable margin]    [base]   [2.0%]
Bear         20%           [lower growth]          [compressed]       [higher] [1.5%]

Scenario Narratives:
  Bull Case:  Favorable macro, market share gains, operating leverage, margin expansion
  Base Case:  Historical trend continuation, modest improvement in line with guidance
  Bear Case:  Competitive pressure, margin compression, macro headwinds, execution risk

Intrinsic Value Output:
  Bull Case IV:    $[value]
  Base Case IV:    $[value]
  Bear Case IV:    $[value]

Probability-Weighted IV = (20% × Bull IV) + (60% × Base IV) + (20% × Bear IV) = $[value]

The probability-weighted IV is the primary output used for investment decision-making.

5×5 Sensitivity Table — Interpretation Guide

Sensitivity Table — Intrinsic Value per Share ($)

              Terminal Growth Rate
WACC     1.0%    1.5%    2.0%    2.5%    3.0%
6.0%     $xxx    $xxx    $xxx    $xxx    $xxx
7.0%     $xxx    $xxx    $xxx    $xxx    $xxx
8.0%     $xxx    $xxx    $xxx    $xxx    $xxx  ← Base Case
9.0%     $xxx    $xxx    $xxx    $xxx    $xxx
10.0%    $xxx    $xxx    $xxx    $xxx    $xxx

[*] Shaded cell = Base Case assumption

Interpretation guide:

  • If the entire table shows a margin of safety vs. current price → high confidence in undervaluation
  • If only a few cells show margin of safety → valuation depends critically on specific assumptions
  • If no cells show margin of safety → stock is expensive under all reasonable DCF scenarios

Common DCF Pitfalls

  1. Garbage in, garbage out: Extrapolating recent high-growth rates too far into the future. Be conservative, especially in years 6–10.
  2. Terminal value dominance: If TV > 70% of enterprise value, the model is more speculation than analysis. Stress-test terminal assumptions aggressively.
  3. WACC too low: Using a WACC below the risk-free rate, or ignoring size/liquidity premiums for smaller companies, artificially inflates intrinsic value.
  4. Ignoring cyclicality: Using peak FCF margins as the base for a cyclical business. Always normalize FCF through a full business cycle.
  5. Ignoring stock-based compensation: SBC is a real, dilutive cost. Subtract it from operating cash flow when calculating true economic FCF.
  6. Single scenario thinking: Presenting only a base case. Always run Bull, Base, and Bear scenarios with explicit assumption differences.
  7. Hidden working capital and capex changes: Rapidly growing companies often consume significant working capital. Ensure this drag is reflected in FCF projections, not hidden in operating cash flow.
  8. Currency and geographic mix: For international businesses, project by geography and apply appropriate discount rates by region.

When DCF Is (and Isn't) the Right Tool

When DCF is most reliable:

  • Stable, mature businesses with predictable, consistent FCF
  • Asset-light businesses with high FCF conversion (e.g., software, consumer brands)
  • Companies with 10+ years of FCF generation history
  • Businesses where future cash flows are reasonably bounded (regulated utilities, subscription SaaS)

When DCF is less reliable (use relative valuation instead):

  • Early-stage growth companies with no positive FCF yet — value lies in future optionality
  • Highly cyclical businesses where normalizing FCF requires significant judgment
  • Financial companies (banks, insurance, REITs) — use Price/Book, Price/Earnings, or dividend discount models instead
  • Companies with lumpy or unpredictable capex cycles
  • Turnaround situations where the path to profitability is uncertain

Input Formats

# Auto-calculate from public financial data
/stock-valuation AAPL

# Specify methods to use
/stock-valuation MSFT --methods dcf,cca,ev-ebitda

# Full multi-method analysis with visual output
/stock-valuation NVDA --full --visual

# Quick single-method valuation
/stock-valuation GOOGL --method dcf --quick

# Custom assumption overrides
/stock-valuation AMZN --growth 15% --wacc 9% --terminal 2.5%

# Compare multiple stocks
/stock-valuation AAPL,MSFT,GOOGL --compare

# With peer set specification
/stock-valuation META --peers SNAP,PINS,TWTR,RDDT

Output Summary

Complete valuation report including:

  1. Business Summary: One-paragraph description of the business model and competitive position
  2. Method Selection Rationale: Why each method was selected or excluded
  3. DCF Analysis: Full 10-year projections, WACC decomposition, sensitivity table (3 scenarios)
  4. Comparable Company Analysis: Peer table with multiples and implied values
  5. Multiple-Based Valuations: EV/EBITDA, P/E, P/FCF implied ranges
  6. Football Field Chart: Visual range of all methods
  7. Composite Intrinsic Value: Weighted average across all methods
  8. Margin of Safety Assessment: Discount/premium to current price
  9. Risk-Adjusted Expected Return: Scenario-weighted expected return calculation
  10. Key Valuation Risks: What assumptions could most materially change the outcome
  11. Analyst Consensus Comparison: How model compares to Street consensus

Thesis Invalidation

After delivering the analysis signal, specify what would reverse it:

If signal is BULLISH — thesis breaks if:

  • intrinsic value declines >20% on updated assumptions OR multiple compression vs. peers worsens
  • [One or two more triggers drawn from this analysis's own drivers, each with a threshold]

If signal is BEARISH — thesis breaks if:

  • all 5 valuation methods show >20% upside at current price simultaneously
  • [One or two more triggers drawn from this analysis's own drivers, each with a threshold]

Re-run this analysis when:

  • Next earnings release
  • Price moves ±15% from current level
  • 60 days have elapsed
  • Material news event (acquisition, leadership change, regulatory decision)

Standard Signal Output

All analysis concludes with this standardized block:

In the output, the box below is followed immediately by this machine-readable footer — the same values as the box, filled in (no placeholders, no comments), so tools can read the result without parsing the box:

{"investskill": "1", "skill": "stock-valuation", "ticker": "<TICKER, or null for a portfolio / market-wide view>", "as_of": "<YYYY-MM-DD from Data & Sources>", "signal": "<BULLISH | NEUTRAL | BEARISH>", "confidence": "<HIGH | MEDIUM | LOW>", "horizon": "<SHORT-TERM | MEDIUM-TERM | LONG-TERM>", "score": 0.0, "action": "<BUY | HOLD | SELL>", "conviction": "<STRONG | MODERATE | WEAK>", "retrieval": "<pasted by user | web/tool retrieval | model memory>", "sources": ["<each primary source, dated>"]}
╔══════════════════════════════════════════════╗
║              INVESTMENT SIGNAL               ║
╠══════════════════════════════════════════════╣
║ Signal:      BULLISH / NEUTRAL / BEARISH     ║
║ Confidence:  HIGH / MEDIUM / LOW             ║
║ Horizon:     SHORT / MEDIUM / LONG-TERM      ║
║ Score:       X.X / 10                        ║
╠══════════════════════════════════════════════╣
║ Action:      BUY / HOLD / SELL               ║
║ Conviction:  STRONG / MODERATE / WEAK        ║
╚══════════════════════════════════════════════╝

Score Guide: 8.0–10.0 Strongly Bullish | 6.0–7.9 Moderately Bullish | 4.0–5.9 Neutral | 2.0–3.9 Moderately Bearish | 0.0–1.9 Strongly Bearish Confidence: HIGH (strong data, clear signals) | MEDIUM (mixed signals) | LOW (limited data, conflicting signals) Horizon: SHORT-TERM (1 week–3 months) | MEDIUM-TERM (3 months–1 year) | LONG-TERM (1+ years)

Disclaimer: Educational analysis only. Not financial advice.

Tax Lens — What You Actually Keep

⚠️ This is not tax advice. Tax law changes, thresholds are indexed, and the right answer depends on facts this skill cannot see (filing status, state, residency, other income, treaty position). Every output of this skill must open and close with: "Educational illustration of how the rules generally work — confirm with a qualified tax professional before acting." If the user asks for a filing position, a specific dollar liability to rely on, or anything about a jurisdiction outside the US mechanics described here, say so and stop.

⚠️ Data Verification — Do This Before Any Analysis

Before running any analysis, always retrieve the latest market data for the ticker:

  1. Fetch current price — use web search or ask the user for the live price, 52-week range, and market cap. Never assume a price from training data.
  2. Confirm key figures — for this skill: each lot's acquisition date, cost basis, and share count; the ex-dividend dates involved; the current tax-year brackets and thresholds (long-term capital-gains brackets, the net investment income tax threshold, the annual capital-loss limit, the non-resident estate-tax exemption). Thresholds are indexed or legislated — state the tax year you are using and ask the user to confirm.
  3. State your data source — fill in the Data & Sources header (next section) so the origin, as-of date, retrieval path, and confidence of every figure are explicit at the top of the output.
  4. Flag stale data explicitly — if live data is unavailable, display this warning before proceeding:

⚠️ Live data unavailable. The following analysis uses training-data estimates which may be significantly out of date. Verify all prices and metrics before making any decisions.

Never silently substitute training-data estimates for current prices. When in doubt, ask the user to paste the latest quote.


📋 Data & Sources Header — Open Every Output With It

Immediately after the not-tax-advice banner (which is always line one), the output carries this provenance block, filled in — never left as placeholders. It is the standard documented on the Data & Accuracy page and the first thing result-validator looks for:

Data & Sources
  As of:      <date the figures represent, e.g. 2026-06-30>
  Source:     <primary docs — SEC EDGAR 10-K/10-Q, company IR, FRED, exchange data …>
  Retrieval:  <pasted by user | web/tool retrieval | model memory>
  Confidence: <HIGH | MEDIUM | LOW>
  • Retrieval: model memory must be paired with Confidence: LOW — memory is a placeholder until confirmed against a primary source.
  • Mixed sources: list each with its own as-of date rather than blending them.
  • Sources disagree on the same figure (a closing price, a deal term): prefer the primary source (SEC filing, company release, exchange data) over an aggregator, then the most recent dated one. Show both values with their sources and dates, say which you used and why, and never average them.
  • Data the user pasted is reported as pasted by user; do not upgrade its confidence beyond what the user's own source supports.

For this skill the Source: line must also name the tax year and the rule sources used (IRS publications 550 / 519 / 559, the Form W-8BEN instructions, the relevant treaty table) — rules, like prices, have an as-of date.


🌐 Output Language — --lang zh-TW

English is the default. With --lang zh-TW — or when the user writes in Traditional Chinese and asks for a Chinese answer — write the whole analysis in Traditional Chinese (繁體中文, Taiwan usage), keeping it machine-checkable:

  • Translate every heading, narrative paragraph, table header and bullet. Give each metric's English name in brackets the first time it appears — 自由現金流(Free Cash Flow, FCF) — so the reader can match it to the glossary and to English analyses.
  • Keep in English tickers, figures and units, SEC form and item names (10-K Item 7, Form 4), the four Data & Sources field labels (As of, Source, Retrieval, Confidence — the Retrieval value may be written in Chinese, e.g. 使用者提供), and the Investment Signal box and JSON footer exactly as specified — field names and the BULLISH / NEUTRAL / BEARISH, HIGH / MEDIUM / LOW, BUY / HOLD / SELL, STRONG / MODERATE / WEAK values. Tools parse them.
  • Title the invalidation section 論點失效條件(Thesis Invalidation), and close with the disclaimer in Chinese: 免責聲明: 僅供教育用途,非投資建議。
  • Use Taiwan financial terms: 殖利率、本益比、營業利益率、自由現金流、聯準會、選擇權、買權/賣權、除息日、預扣稅.

Overview

Every other framework in the catalog stops at the pre-tax number. stock-eval says whether to own it, stock-valuation what it is worth, position-ladder how to build the position, dividend-analysis how safe the payout is. None of them answers the question that decides what the investor keeps: what does this trade or this portfolio cost in tax, and is there a cheaper way to do the same thing?

This skill is that layer. Given a position, a proposed trade, or a whole portfolio with lots and dates, it works through the US mechanics that move the after-tax result — holding period, wash-sale windows, the qualified-dividend test, which lot to sell, whether a loss is worth harvesting and what to swap into, which account each asset belongs in — and estimates the annual tax drag the portfolio is carrying. The --non-us module switches to the rules that apply to a non-resident alien holding US stocks through a US or international broker, which differ completely from what US-centric content teaches and are almost never covered: withholding at source, the treaty rate (or the lack of a treaty), how capital gains are generally treated, the estate-tax exposure that starts far lower than for US persons, and the Irish-domiciled UCITS route many non-US investors use instead.

What it is not. It is not a return calculator, a filing tool, or a substitute for a professional. It does not cover state taxes, the alternative minimum tax, employer stock plans (ISOs, ESPPs, RSUs) beyond flagging them, or the investor's home-country taxation in --non-us mode. It names the mechanism, shows the arithmetic on the user's own numbers, and says where the professional's judgment has to take over.


1. Modes

ModeInvocationScope
Tradetax-lens AVGO — sell 20 sh bought 2026-05-30 @ $128, now $122One proposed sale, purchase, or swap: character of the gain/loss, wash-sale check, lot choice, timing alternatives
Positiontax-lens AVGO + lotsEvery lot of one holding: unrealized gain/loss by lot and character, harvest candidates, when short-term lots turn long-term
Portfoliotax-lens --portfolio + holdings with lotsWhole-portfolio view: annual tax drag, account placement, harvest pairs, concentration of embedded gains
Non-USany of the above with --non-us [country]Same questions for a non-resident alien: withholding, treaty rate, capital-gains treatment, estate-tax exposure, UCITS alternatives

2. Inputs

InputRequiredDefault if unstated
Ticker(s)✅—
Lots: shares, acquisition date, cost basis per lot✅ for Trade / Position / Portfolioask — without dates no holding-period or wash-sale answer is possible
Proposed action (sell / buy / swap, quantity, date)✅ for Trade—
Account type per holding (taxable, traditional IRA / 401(k), Roth)recommendedtaxable
Filing status and approximate marginal bracketrecommended for US modeillustrate at the 15% long-term / 24% ordinary bracket and say so
Recent sales and purchases in the same or substantially identical security (±30 days, all accounts)required for the wash-sale checkask
Country of tax residence, W-8BEN on file (yes/no), broker domicilerequired for --non-usask
Days physically present in the US during the tax yearrequired for the --non-us capital-gains answerif unknown, report the capital-gains treatment as indeterminate rather than assuming under 183
Tax yearrecommendedcurrent calendar year, stated

Never assume a bracket silently. If the user does not give one, run the illustration at a stated default and show how the answer changes one bracket up and down.


3. Framework — US mode

Phase 1 — Character of the gain or loss

For each lot involved:

TestRule (general)What to output
Holding periodHeld more than one year → long-term; one year or less → short-term. The clock starts the day after the trade date and includes the sale dateDays held, the date each short-term lot turns long-term, and the dollar difference between selling now and selling on that date at today's price
Long-term ratePreferential brackets (generally 0% / 15% / 20% by taxable income)Which bracket the user's stated income lands in
Short-term rateTaxed as ordinary income at the marginal bracketThe marginal rate the user stated
Net investment income taxAn additional 3.8% on investment income above a modified-AGI threshold (not indexed; higher for joint filers)Whether the user is plausibly above it; if unknown, show both
NettingShort-term gains and losses net first, long-term net separately, then the two net against each other; up to $3,000 of net capital loss offsets ordinary income per year ($1,500 for married filing separately), the rest carries forward indefinitelyThe user's net position after the proposed trade

"Wait N days" is the cheapest tax strategy there is. Always compute it: if a lot turns long-term within ~60 days, show the tax saved at today's price and the price fall that would wipe out that saving — so the reader can weigh a tax deferral against market risk instead of ignoring one of them.

Phase 2 — Wash-sale window check

The rule, in general terms: a loss is disallowed if the same or a substantially identical security is bought within 30 days before or after the sale (a 61-day window, counting the sale date). The disallowed loss is not gone — it is added to the basis of the replacement shares and the holding period carries over. Gains are never washed.

Check, and state each result explicitly:

  1. Any purchase of the same ticker in the 30 days before the loss sale (including dividend reinvestment — DRIP purchases are the classic accidental wash).
  2. Any planned or automatic purchase in the 30 days after (DRIP, a position-ladder rung, an options assignment).
  3. Purchases in other accounts — the rule applies across the investor's accounts, and a repurchase inside an IRA is treated as permanently losing the loss (the basis adjustment has nowhere to go).
  4. Substantially identical: same ticker, options on it, convertible into it — yes. A different company in the same industry — no. Two index funds tracking the same index — unsettled and conservative practice treats them as risky; two funds tracking different indexes — the common practice for harvesting.

Output a 61-day timeline with the sale in the middle and every purchase marked ✓ safe / ✗ wash / ⚠️ scheduled.

Phase 3 — Dividends: qualified or not

QuestionGeneral rule
Is the payer eligible?US corporations and qualified foreign corporations — yes. REITs, most MLPs, money-market and bond-fund income — generally not qualified (REIT ordinary dividends may qualify for a separate 20% deduction instead)
Did the holder meet the holding period?The stock must be held more than 60 days during the 121-day period that begins 60 days before the ex-dividend date. Hedged shares (deep puts, short calls) can fail the test
What is the consequence?Qualified → long-term capital-gains rates. Not qualified → ordinary income

For a dividend payer, show the ex-dividend dates in the period, whether each lot passes the test, and the after-tax yield at the user's bracket — that is the number dividend-analysis cannot give.

Phase 4 — Which lot to sell

MethodWhen it helpsCaveat
Specific identificationAlmost always — choose the lot that produces the character and size of gain/loss you wantMust be designated at or before the trade, in the form the broker accepts; a retroactive choice is not allowed
FIFO (default at most brokers)Long-held, low-basis lots first — largest gainsOften the worst choice for a partial sale in a rising stock
Highest cost first (HIFO)Minimizes the current gain or maximizes the harvested lossMay realize short-term gains; check character, not just size
Lowest cost firstDeliberately realizing gains in a 0% bracket year, or before a rate changeRare, but real

Produce the lot table: lot · shares · basis · gain/loss · character · days to long-term · recommended for this trade? — and the total tax at the user's bracket under FIFO vs. the recommended specific lots.

Phase 5 — Tax-loss harvesting pairs

For every lot with an unrealized loss above a materiality floor (state it, e.g. the larger of $500 or 5% of the lot):

  • Value of the harvest = loss × the rate it offsets (short-term losses are worth more when they offset short-term gains or ordinary income).
  • Replacement: name a similar-but-not-substantially-identical holding that keeps the exposure for the 31 days (a different index or a peer, never the same ticker or an option on it). Say what the tracking difference costs.
  • The reversal: after 31 days, whether to swap back (a new taxable event) or keep the replacement.
  • Basis reset warning: harvesting lowers basis — the tax is deferred, not eliminated, unless the position is held until a step-up or a donation.

Phase 6 — Account placement

General placement logic, shown as a table of the user's holdings:

Asset behaviourTaxable accountTax-deferred (traditional IRA / 401(k))Tax-free (Roth)
Broad, low-turnover equity index; individual stocks held long✅ best — long-term rates, harvesting possible, step-upfinefine
Bonds, REITs, high-yield, high-turnover active funds✗ ordinary income every year✅✅
Highest expected growthokok✅ best — growth never taxed
Foreign equity funds with withholding✅ (the foreign tax credit is only available here)✗ credit lost✗ credit lost

Flag every holding sitting in the wrong column and estimate the annual cost of leaving it there.

Phase 7 — Annual tax drag

Tax drag (% of portfolio per year) ≈
    Σ weight_i × [ yield_i × ( qualified_i × LTCG rate + (1 − qualified_i) × ordinary rate ) ]
  + realized-gain turnover × blended gain rate
  + non-qualified income (bond interest, REIT) × ordinary rate

Scope — the terms are disjoint: yield_i is the dividend yield of the *equity* holdings only
(stocks and equity funds), split into its qualified and non-qualified parts; the third line
covers income that is never a qualified dividend (bond and money-market interest, REIT and
MLP distributions), so it is not counted in yield_i as well.

Report the drag in percentage points and dollars, the three largest contributors, and the drag after the Phase 5–6 changes. State the assumed rates in the table, not in prose.


4. Framework — --non-us module (non-resident alien holding US stocks)

The mechanics below describe the US side only. The investor's country of residence taxes the same income under its own rules, and that is out of scope — say so in the output.

Phase 1 — Status and paperwork

  • W-8BEN certifies non-US status and claims any treaty rate. It is generally valid until the end of the third calendar year after signing; an expired form means the broker withholds at the statutory rate (and may apply backup withholding). Output: on file? expiry year? treaty article claimed?
  • Form 1042-S is the year-end statement of US-source income and tax withheld — the document to reconcile against, and often the one the home-country return needs.
  • If the user says they are a US citizen or green-card holder living abroad, stop the --non-us module: they are taxed as US persons worldwide, and non-US funds (UCITS) raise PFIC issues for them. Point back to US mode.

Phase 2 — Dividends: withholding and the treaty rate

ItemGeneral ruleOutput
Statutory withholding on US-source dividends30%, withheld by the broker or paying agent at sourceEffective yield after withholding
Treaty rateReduced (commonly 15%, some treaties 10% or lower) only if the residence country has an income-tax treaty with the US and a valid W-8BEN claims itName the treaty and the article, or state clearly that no treaty exists (for example Taiwan, Hong Kong, Singapore have no US income-tax treaty, so 30% applies)
Fund distributionsOrdinary dividends withheld as above; the treatment of short-term / long-term capital-gain and return-of-capital distributions differs — reconcile against the 1042-S rather than assumeFlag distribution types
InterestPortfolio interest (Treasuries, most corporate bonds) is generally exempt from US withholdingNote where relevant
Derivatives on US equitiesDividend-equivalent payments on certain swaps and deep-in-the-money options can be withheld as if dividends (Section 871(m))Flag if the user trades options on dividend payers

After-withholding yield is the headline number: a 3.0% dividend yield is 2.1% to a no-treaty holder and 2.55% at a 15% treaty rate. dividend-analysis reports the gross figure; this section converts it.

Phase 3 — Capital gains

US capital gains on stocks are generally not taxed by the US for a non-resident alien. State the general conditions and the exceptions rather than a blanket promise:

  • The general exemption applies when the individual is not present in the US for 183 days or more in the tax year and the gain is not effectively connected with a US trade or business. The day count is an input, not an assumption — if the user has not supplied it, state the treatment as indeterminate and show both outcomes.
  • Exceptions: gains effectively connected with a US business; gains on US real property interests (including certain REITs) under FIRPTA, which are taxed and often withheld; and any home-country tax on the same gain, which is the rule that actually matters for most investors.
  • Losses cannot be used against US tax the investor does not owe — the harvesting logic of US mode is irrelevant here except for the home-country return. Say so; do not port Phase 5 across.

Phase 4 — US estate-tax exposure

This is the item most non-US investors have never heard of, and the one with the largest tail:

  • US-situs assets of a non-resident alien — US-domiciled stocks and US-domiciled ETFs count; US Treasury bonds, bank deposits, and shares of non-US companies (including ADRs of foreign issuers) generally do not — are subject to US estate tax above an exemption of only $60,000 (not the multi-million exemption US persons have), at rates that reach 40%, unless an estate-tax treaty provides otherwise (only a minority of countries have one; Taiwan, Hong Kong, and Singapore do not).
  • Output: the user's US-situs total today, the excess over $60,000, an illustrative exposure at the statutory brackets, and the two standard mitigations — holding the same exposure through non-US-domiciled funds (Phase 5) or estate planning structures that are entirely a professional's domain.
  • Brokers may freeze the account pending a US estate-tax clearance; mention it as a practical consequence, not as advice.

Phase 5 — Irish-domiciled UCITS alternatives

The common route for non-US investors, with an honest cost/benefit table:

DimensionUS-domiciled ETF (e.g. an S&P 500 ETF listed in New York)Irish-domiciled UCITS ETF (same index, listed in London / Amsterdam / Frankfurt)
Dividend withholding30% (or treaty rate) at the investor level15% at the fund level under the US–Ireland treaty; no further Irish withholding on distributions to non-Irish holders
Accumulating share classNot available (US funds must distribute)Available — reinvests inside the fund; whether the home country taxes the notional income is a home-country question
US estate taxUS-situs — exposed above $60,000Not US-situs — outside the US estate-tax net
Expense ratio and spreadUsually lower TER, tighter spreads, deepest liquidityTypically a few basis points higher TER and wider spreads; some are large and liquid, many are not
TrackingBenchmarkSlightly lower drag from the 15% fund-level withholding vs. 30% at investor level — for a no-treaty investor the UCITS fund often wins after tax despite the higher TER
AccessAny brokerNeeds a broker with European exchange access; US brokers generally will not sell UCITS to anyone, and US persons should not buy them (PFIC)

Show the arithmetic for the user's own case: gross yield → after-withholding yield under both routes → net of the TER difference → the estate-tax exposure difference. Hand the result to etf-analysis for the fund-level comparison.


5. Scoring — Tax Efficiency Score (0–10)

The headline number measures how tax-efficient the trade or portfolio is as specified, not whether the investment is good.

ComponentWeight10 looks like0 looks like
Character & timing30%Gains long-term, losses short-term; no lot turns long-term within 60 days of a proposed short-term saleLarge short-term gains realized days before turning long-term
Wash-sale hygiene20%No purchases in the 61-day window across any accountA harvested loss disallowed by a DRIP or IRA repurchase
Lot selection15%Specific-ID chosen and designatedFIFO by default on a partial sale of a long-held winner
Account placement15%Every holding in the right columnBond funds and REITs in taxable, growth in traditional IRA
Harvest capture10%Material losses harvested into non-identical replacementsLosses left on the table in a high-gain year
Drag & paperwork10%Drag below ~0.5% / yr; W-8BEN current and treaty claimed (non-US)Drag > 1.5% / yr; expired W-8BEN, statutory withholding, unaddressed estate exposure

In --non-us mode, replace "Harvest capture" with Estate-tax exposure (10 looks like: US-situs holdings below the exemption or moved to non-US-domiciled funds; 0: a large US-situs portfolio with no plan).


6. Output Format

  1. Not-tax-advice banner (verbatim, first line)
  2. Data & Sources header — including the tax year and rule sources
  3. Executive summary — three sentences: the tax cost of the action as specified, the cheapest alternative, and the one thing to confirm with a professional
  4. Lot table (Trade / Position / Portfolio) — lot · shares · basis · gain/loss · character · days to long-term · recommended?
  5. Wash-sale timeline — 61-day window with every purchase marked
  6. Dividend test — ex-dates, holding-period pass/fail, after-tax yield (or after-withholding yield in --non-us)
  7. Alternatives table — "as specified" vs. "wait N days" vs. "specific lots" vs. "harvest & replace", each with estimated tax and market-risk trade-off
  8. Placement & drag (Portfolio mode) — misplaced holdings, drag before / after
  9. --non-us block — W-8BEN status, treaty or no-treaty rate, capital-gains treatment with exceptions, estate-tax exposure, UCITS comparison
  10. Thesis Invalidation — what would change this analysis (below)
  11. Investment Signal block — the standard box (below)
  12. Not-tax-advice banner (verbatim, last line)

Example

User: tax-lens AVGO — I hold 20 sh bought 2026-05-30 @ $128 (taxable) and 10 sh bought
      2025-05-14 @ $96. Price $122. I want to sell 15 shares and rebuy lower with the
      position-ladder plan. 24% bracket, single, DRIP is on.

The assistant opens with the not-tax-advice banner and a Data & Sources header (tax year 2026,
IRS Pub 550, prices pasted by user). Lot table: the 2026 lot shows a $6/sh short-term loss;
the 2025 lot a $26/sh long-term gain that has already passed one year. Selling 15 from the
2026 lot by specific identification realizes a $90 short-term loss; FIFO would instead sell
the 10 long-term gain shares first (+$260 gain, ~$39 tax at 15%) plus 5 loss shares. The
wash-sale timeline flags two problems: DRIP reinvested a dividend on 2026-06-30 (inside the
window — part of the loss is disallowed and added to those shares' basis) and every
position-ladder rung inside the next 30 days would wash the rest. Alternatives: turn DRIP
off and place the first rung on day 31; or accept the wash (the loss moves into the basis of
the re-bought shares and is not lost, only deferred). Tax Efficiency Score 4.5 → NEUTRAL:
the plan is fine once the DRIP and rung timing are fixed. Closes with the banner.

User: tax-lens --non-us Taiwan — I hold $180k of VOO and $40k of AAPL at a US broker,
      W-8BEN signed 2023.

The assistant states that Taiwan has no US income-tax treaty, so dividends are withheld at 30%
(VOO's ~1.3% gross yield → ~0.9% net); the W-8BEN signed in 2023 expires at the end of 2026
and must be renewed or withholding may rise further; US capital gains are generally not taxed
by the US for a non-resident (conditions and exceptions stated) but Taiwan's own rules apply;
US-situs assets total $220k, $160k above the $60k exemption — an illustrative estate-tax
exposure it quantifies at the statutory brackets — and shows the Irish-UCITS comparison for
the VOO sleeve (15% fund-level withholding, no US estate exposure, ~0.03–0.05% higher TER).
Score 3.5 → BEARISH on tax efficiency as held; the investment view is unchanged. Recommends a
cross-border professional for the estate question. Closes with the banner.

Notes

  • Rules move. Brackets and thresholds are indexed annually; the wash-sale rule, the qualified-dividend test, and the treatment of non-resident capital gains have all been the subject of proposals. State the tax year, cite the publication, and tell the user to confirm.
  • Never compute a liability to be relied upon. Show the mechanics on the user's numbers as an illustration and label it as such.
  • Out of scope, always flagged: state income tax, AMT, employer equity (ISO/ESPP/RSU) beyond identifying it, trusts and entities, the home-country side of --non-us, and gifting/inheritance planning.
  • Pairs with: position-ladder (rung timing vs. the wash window; the ceiling vs. concentration of embedded gains), portfolio-review (placement), dividend-analysis (gross yield in, after-tax yield out), etf-analysis (US vs. UCITS fund comparison), risk-stress-test (what a forced de-risking would realize).
  • For the zh-TW audience: Taiwan has no US income-tax or estate-tax treaty; the 30% withholding, the $60,000 estate exemption, and the UCITS route are the three facts that change the arithmetic most. Home-country reporting of overseas income is a Taiwan tax question this skill does not answer.

Thesis Invalidation

After delivering the analysis signal, specify what would reverse it:

If signal is BULLISH (tax-efficient as specified) — the assessment breaks if:

  • A purchase of the same security lands inside the 61-day window (DRIP, a ladder rung, an assignment) and washes a loss the plan relied on
  • The user's bracket, filing status, or residency changes — every rate in the table moves with it
  • A rule the analysis leaned on changes (rate brackets, the wash-sale scope, treaty status, the estate exemption) — re-run against the new tax year

If signal is BEARISH (tax-inefficient as specified) — the assessment breaks if:

  • The market moves enough that the tax saving from waiting is smaller than the price risk of waiting — the "wait N days" table shows the break-even fall; below it, sell
  • The user has offsetting losses or a low-income year that changes the character math (a 0% long-term bracket year makes realizing gains desirable)
  • The broker confirms specific-ID designation or a treaty rate the analysis assumed was unavailable

Re-run this analysis when:

  • A short-term lot is within 60 days of turning long-term
  • Before any sale at a loss, and 31 days after it (the window closes)
  • Each ex-dividend date for holdings where the qualified test is close
  • Year-end (harvest window) and the new tax year (indexed thresholds)
  • W-8BEN expiry year, a change of residence, or a change in treaty status (--non-us)

Standard Signal Output

This skill measures tax efficiency of the action or portfolio as specified, not the merit of the investment, so state the mapping: Signal: BULLISH means proceed as planned from a tax standpoint; NEUTRAL means the same trade should be done differently (lot choice, timing, account); BEARISH means the action as specified is tax-inefficient (a wash, a short-term gain days from long-term, a large avoidable withholding or estate exposure). Action: BUY / HOLD / SELL refers to the tax plan — adopt it / adjust it / abandon it — never to the stock. Read the investment view from stock-eval, dividend-analysis, or etf-analysis.

All analysis concludes with this standardized block:

In the output, the box below is followed immediately by this machine-readable footer — the same values as the box, filled in (no placeholders, no comments), so tools can read the result without parsing the box:

{"investskill": "1", "skill": "tax-lens", "ticker": "<TICKER, or null for a portfolio / market-wide view>", "as_of": "<YYYY-MM-DD from Data & Sources>", "signal": "<BULLISH | NEUTRAL | BEARISH>", "confidence": "<HIGH | MEDIUM | LOW>", "horizon": "<SHORT-TERM | MEDIUM-TERM | LONG-TERM>", "score": 0.0, "action": "<BUY | HOLD | SELL>", "conviction": "<STRONG | MODERATE | WEAK>", "retrieval": "<pasted by user | web/tool retrieval | model memory>", "sources": ["<each primary source, dated>"]}
╔══════════════════════════════════════════════╗
║              INVESTMENT SIGNAL               ║
╠══════════════════════════════════════════════╣
║ Signal:      BULLISH / NEUTRAL / BEARISH     ║
║ Confidence:  HIGH / MEDIUM / LOW             ║
║ Horizon:     SHORT / MEDIUM / LONG-TERM      ║
║ Score:       X.X / 10                        ║
╠══════════════════════════════════════════════╣
║ Action:      BUY / HOLD / SELL               ║
║ Conviction:  STRONG / MODERATE / WEAK        ║
╚══════════════════════════════════════════════╝

Score Guide: 8.0–10.0 Strongly Bullish | 6.0–7.9 Moderately Bullish | 4.0–5.9 Neutral | 2.0–3.9 Moderately Bearish | 0.0–1.9 Strongly Bearish Confidence: HIGH (strong data, clear signals) | MEDIUM (mixed signals) | LOW (limited data, conflicting signals) Horizon: SHORT-TERM (1 week–3 months) | MEDIUM-TERM (3 months–1 year) | LONG-TERM (1+ years)

Note: The Score above is the Tax Efficiency Score, mapped onto the standard scale for cross-skill comparability. It rates how much of the pre-tax result the plan keeps — not how attractive the security is. Confidence is capped at MEDIUM whenever the user's bracket, residency, or lot dates were assumed rather than supplied.

Disclaimer: Educational analysis only. Not financial advice, and not tax advice — an illustration of how the rules generally work, for a stated tax year, on the user's own numbers. Tax law changes and depends on facts this skill cannot see. Confirm with a qualified tax professional before acting.

Technical Analysis

⚠️ Data Verification — Do This Before Any Analysis

Before running any analysis, always retrieve the latest market data for the ticker:

  1. Fetch current price — use web search or ask the user for the live price, 52-week range, and market cap. Never assume a price from training data.
  2. Confirm key figures — the price and volume series (interval, date range, split/dividend adjustment) the indicators are computed from, the latest close and its date, and the benchmark/sector ETF used for relative strength.
  3. State your data source — fill in the Data & Sources header (next section) so the origin, as-of date, retrieval path, and confidence of every figure are explicit at the top of the output.
  4. Flag stale data explicitly — if live data is unavailable, display this warning before proceeding:

⚠️ Live data unavailable. The following analysis uses training-data estimates which may be significantly out of date. Verify all prices and metrics before making any decisions.

Never silently substitute training-data estimates for current prices. When in doubt, ask the user to paste the latest quote.


📋 Data & Sources Header — Open Every Output With It

The first thing in the output is this provenance block, filled in — never left as placeholders. It is the standard documented on the Data & Accuracy page and the first thing result-validator looks for:

Data & Sources
  As of:      <date the figures represent, e.g. 2026-06-30>
  Source:     <primary docs — SEC EDGAR 10-K/10-Q, company IR, FRED, exchange data …>
  Retrieval:  <pasted by user | web/tool retrieval | model memory>
  Confidence: <HIGH | MEDIUM | LOW>
  • Retrieval: model memory must be paired with Confidence: LOW — memory is a placeholder until confirmed against a primary source.
  • Mixed sources: list each with its own as-of date rather than blending them.
  • Sources disagree on the same figure (a closing price, a deal term): prefer the primary source (SEC filing, company release, exchange data) over an aggregator, then the most recent dated one. Show both values with their sources and dates, say which you used and why, and never average them.
  • Data the user pasted is reported as pasted by user; do not upgrade its confidence beyond what the user's own source supports.

🌐 Output Language — --lang zh-TW

English is the default. With --lang zh-TW — or when the user writes in Traditional Chinese and asks for a Chinese answer — write the whole analysis in Traditional Chinese (繁體中文, Taiwan usage), keeping it machine-checkable:

  • Translate every heading, narrative paragraph, table header and bullet. Give each metric's English name in brackets the first time it appears — 自由現金流(Free Cash Flow, FCF) — so the reader can match it to the glossary and to English analyses.
  • Keep in English tickers, figures and units, SEC form and item names (10-K Item 7, Form 4), the four Data & Sources field labels (As of, Source, Retrieval, Confidence — the Retrieval value may be written in Chinese, e.g. 使用者提供), and the Investment Signal box and JSON footer exactly as specified — field names and the BULLISH / NEUTRAL / BEARISH, HIGH / MEDIUM / LOW, BUY / HOLD / SELL, STRONG / MODERATE / WEAK values. Tools parse them.
  • Title the invalidation section 論點失效條件(Thesis Invalidation), and close with the disclaimer in Chinese: 免責聲明: 僅供教育用途,非投資建議。
  • Use Taiwan financial terms: 殖利率、本益比、營業利益率、自由現金流、聯準會、選擇權、買權/賣權、除息日、預扣稅.

Perform technical analysis of US stocks using price action, chart patterns, and technical indicators.

Chart Pattern Analysis

  1. Trend Identification

    • Primary trend (uptrend, downtrend, sideways)
    • Trend strength and momentum
    • Support and resistance levels
    • Trendline analysis
  2. Classic Chart Patterns

    • Head and shoulders, inverse H&S
    • Double/triple tops and bottoms
    • Cup and handle
    • Triangles (ascending, descending, symmetrical)
    • Flags and pennants
    • Wedges and channels
  3. Candlestick Patterns

    • Reversal patterns (doji, hammer, shooting star, engulfing)
    • Continuation patterns
    • Multi-candle formations

Technical Indicators

  1. Trend Indicators

    • Moving averages (SMA, EMA: 20, 50, 200-day)
    • Moving average crossovers
    • MACD (Moving Average Convergence Divergence)
    • ADX (Average Directional Index)
  2. Momentum Indicators

    • RSI (Relative Strength Index)
    • Stochastic oscillator
    • Williams %R
    • Rate of Change (ROC)
  3. Volume Indicators

    • Volume trends and patterns
    • On-Balance Volume (OBV)
    • Volume-weighted average price (VWAP)
    • Accumulation/Distribution line
  4. Volatility Indicators

    • Bollinger Bands
    • Average True Range (ATR)
    • Volatility Index correlation

Price Levels

  • Key support and resistance zones
  • Fibonacci retracement levels
  • Pivot points
  • Round number psychology
  • Gap analysis

Market Context

  • Relative strength vs. market indices
  • Sector rotation analysis
  • Market breadth indicators
  • Correlation with related assets

Moving Average Chart Analysis

When given a stock ticker, always generate an MA chart and trade recommendation table as part of the analysis. This is the first required output section.

MA Chart — Markdown Format

Retrieve or estimate recent price data and compute the following MAs:

MA PeriodLabelTrend Role
30-day SMAMA30Short-term trend
60-day SMAMA60Intermediate trend
90-day SMAMA90Medium-term trend
200-day SMAMA200Long-term bull/bear line
365-day SMAMA365Annual trend baseline

Required output — MA Position Table:

━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
  MOVING AVERAGE CHART — [TICKER]          [DATE]
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
  Current Price : $[PRICE]

  MA       Value      vs Price    Signal
  ──────   ────────   ─────────   ───────────────────
  MA30     $[val]     ▲ +X.X%     [Above/Below] ← Short-term
  MA60     $[val]     ▲ +X.X%     [Above/Below] ← Intermediate
  MA90     $[val]     ▲ +X.X%     [Above/Below] ← Medium-term
  MA200    $[val]     ▼ -X.X%     [Above/Below] ← Long-term bull/bear
  MA365    $[val]     ▼ -X.X%     [Above/Below] ← Annual baseline
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
  MA Stack: [BULLISH / BEARISH / MIXED]
  (Bullish = MA30 > MA60 > MA90 > MA200 > MA365)
━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━

ASCII Trend Chart (last ~90 trading days):

Price ($)
[HIGH] ┤         ╭──╮    ╭──╮
       ┤    ╭────╯  ╰────╯  ╰──╮       ← Price
       ┤────╯                   ╰──╮
       ┤·  ·  ·  ·  ·  ·  ·  ·  · ╰──  ← MA30
       ┤══════════════════════════════   ← MA60
       ┤- - - - - - - - - - - - - - -   ← MA90
[LOW]  ┤══════════════════════════════   ← MA200
       └─┬─────┬─────┬─────┬─────┬──→
       -90d  -70d  -50d  -30d  -10d  Today

Legend: ──── Price  · · · MA30  ════ MA60/MA200  - - - MA90

MA Crossover Signals

Look for and report these high-significance crossover events. The standard data set is MA30 / MA60 / MA90 / MA200 / MA365 (the position table above); MA50 is computed in addition, solely because the classic 50/200 cross is what the market watches — state its value when you report it:

CrossoverTypeSignificance
MA30 crosses above MA60Golden Cross (short)Short-term bullish momentum
MA30 crosses below MA60Death Cross (short)Short-term bearish reversal
MA50 crosses above MA200Golden Cross (classic)Major long-term bullish signal
MA50 crosses below MA200Death Cross (classic)Major long-term bearish signal
Price reclaims MA200 from belowBreakoutShift from bear to bull regime
Price loses MA200 supportBreakdownShift from bull to bear regime

State: most recent crossover event + how many days ago it occurred.


Trade Recommendation from MA Analysis

After computing MAs, output a clear, specific trade recommendation table:

╔══════════════════════════════════════════════════════════════════╗
║              MA-BASED TRADE RECOMMENDATION — [TICKER]           ║
╠══════════════════════════════════════════════════════════════════╣
║  Current Price : $[PRICE]                                       ║
╠══════════════════════════════════════════════════════════════════╣
║  OPERATION      │ ENTRY PRICE   │ TARGET        │ STOP-LOSS     ║
║  ───────────────┼───────────────┼───────────────┼────────────── ║
║  [BUY/SELL/HOLD]│ $[price]      │ $[price]      │ $[price]      ║
╠══════════════════════════════════════════════════════════════════╣
║  Rationale: [1-sentence MA-based reason]                        ║
║  Basis:     MA[X] = $[val] acting as [support/resistance]       ║
║  R/R Ratio: 1 : [X.X]                                          ║
║  Horizon:   [SHORT / SWING / LONG-TERM]                         ║
╚══════════════════════════════════════════════════════════════════╝

Examples of well-formed recommendations:

  • "$185 is a good BUY entry for AAPL — price pulled back to MA90 ($183) which held as support, targeting MA resistance at $198 with stop below MA200 ($175). R/R 1:2.1."
  • "$310 is a good SELL/SHORT entry for TSLA — price rejected MA60 ($312) as resistance while MA30 crossed below MA60, targeting $275 with stop at $320. R/R 1:2.5."
  • "HOLD for MSFT — price is compressing between MA30 ($415) and MA60 ($420); wait for a confirmed break above MA60 before adding."

Decision rules:

  • Price above all 5 MAs in bullish stack → BUY on MA30/MA60 dips
  • Price below MA200 but reclaiming MA90 → speculative BUY with tight stop
  • Price below all 5 MAs → SELL/AVOID; short only if MA30 < MA60 confirmed
  • Price between MAs with no clear stack → HOLD, wait for resolution

Chart Visualization Support

When --chart flag is used, include chart specifications and data tables:

1. Price Chart with Moving Averages

Chart Type: Candlestick chart with overlay lines Data Table:

Date        Open      High      Low       Close     Volume      SMA_30    SMA_60    SMA_200
2024-01-01  [price]   [price]   [price]   [price]   [volume]    [price]   [price]   [price]
2024-01-02  [price]   [price]   [price]   [price]   [volume]    [price]   [price]   [price]
...
[Last 90 trading days for detailed view]

Visual Specifications:

  • Candlesticks: Green for up days, red for down days
  • SMA 30: Blue line (short-term trend)
  • SMA 60: Orange line (intermediate trend)
  • SMA 200: Purple line (long-term trend)
  • Support/resistance: Horizontal dashed lines with labels

2. Volume Bars

Chart Type: Bar chart (synchronized with price chart) Data Table:

Date        Volume      Avg_Volume_20    Above_Avg    Color
2024-01-01  [volume]    [volume]         [Yes/No]     [Green/Red]
2024-01-02  [volume]    [volume]         [Yes/No]     [Green/Red]
...

Visual Specifications:

  • Green bars: Volume on up days
  • Red bars: Volume on down days
  • Dashed line: 20-day average volume
  • Highlight bars exceeding 2x average volume

3. RSI Indicator Panel

Chart Type: Line chart with overbought/oversold zones Data Table:

Date        RSI_14    Overbought_70    Oversold_30    Signal
2024-01-01  [value]   70               30             [Neutral/OB/OS]
2024-01-02  [value]   70               30             [Neutral/OB/OS]
...

Visual Specifications:

  • RSI line: Blue
  • Overbought zone (>70): Red shaded area
  • Oversold zone (<30): Green shaded area
  • Midline (50): Dashed gray line
  • Divergence annotations if present

4. MACD Indicator Panel

Chart Type: Line chart with histogram Data Table:

Date        MACD      Signal    Histogram    Crossover
2024-01-01  [value]   [value]   [value]      [None/Bullish/Bearish]
2024-01-02  [value]   [value]   [value]      [None/Bullish/Bearish]
...

Visual Specifications:

  • MACD line: Blue
  • Signal line: Red
  • Histogram: Green (positive), Red (negative)
  • Zero line: Black dashed
  • Crossover markers: Arrows or dots

5. Bollinger Bands

Chart Type: Overlay on price chart Data Table:

Date        Price     BB_Upper    BB_Middle    BB_Lower    %B       Bandwidth
2024-01-01  [price]   [price]     [price]      [price]     [value]  [value]
2024-01-02  [price]   [price]     [price]      [price]     [value]  [value]
...

Visual Specifications:

  • Upper band: Red dashed line
  • Middle band (20 SMA): Blue solid line
  • Lower band: Green dashed line
  • Shaded area between bands: Light gray
  • Price line: Black candlesticks

6. Support & Resistance Levels

Annotation Overlay:

Level Type       Price      Strength    Touches    Last Test    Notes
Resistance       $[price]   Strong      5          [date]       Historical high
Resistance       $[price]   Medium      3          [date]       Recent rejection
Support          $[price]   Strong      4          [date]       Key accumulation zone
Support          $[price]   Weak        2          [date]       Minor bounce level
Fibonacci 61.8%  $[price]   N/A         -          -            Retracement level

7. Pattern Recognition Annotations

Chart Overlays:

Pattern Type              Start Date    End Date      Status        Target        Risk/Reward
Head & Shoulders          2024-01-15    2024-02-10    Confirmed     $[price]      1:3.2
Ascending Triangle        2024-02-01    Current       Forming       $[price]      1:2.5
Flag Pattern             2024-02-05    2024-02-08    Complete      $[price]      1:2.8

Visual Elements:

  • Pattern boundaries: Colored rectangles or polygons
  • Trendlines: Dashed connecting lines
  • Breakout points: Arrows indicating direction
  • Target levels: Horizontal lines with labels

ASCII Chart Example (Terminal Display)

Technical Chart: AAPL (Last 30 Days)

Price ($)
180 ┤                                    ╭─╮
175 ┤                              ╭─────╯ ╰─╮
170 ┤                        ╭─────╯         ╰──╮
165 ┤                  ╭─────╯                  ╰─╮  ← Current: $172.50
160 ┤            ╭─────╯                          ╰──
155 ┤      ╭─────╯                                      SMA 30: $168.23
150 ┤──────╯                                            SMA 60: $165.80
    └──┬─────┬─────┬─────┬─────┬─────┬─────┬──          ══════ Resistance: $178
     1/15  1/22  1/29  2/05  2/12  2/19  2/26          ══════ Support: $155

RSI (14)
 70 ┤                                    ● ← 68 (approaching OB)
 50 ┤                        ●     ●
 30 ┤            ●
    └──┬─────┬─────┬─────┬─────┬─────┬─────┬──

Volume (M)
100M┤                                    █
 75M┤                              █     █
 50M┤            █     █     █     █  █  █  █
 25M┤      █  █  █  █  █  █  █  █  █  █  █  █
    └──┬─────┬─────┬─────┬─────┬─────┬─────┬──

Signals:
↗ Bullish: Price above SMA 30 & 60
● RSI approaching overbought (68) - potential pullback
📊 Volume increasing on up days - healthy uptrend
🎯 Target: $178 (resistance), Stop: $165 (below SMA 60)

Integration with Report Generator

When --chart flag is used, output is optimized for /report-generator:

  • All chart data tables formatted for Chart.js/Plotly
  • Color schemes specified for consistent visualization
  • Annotations and overlays clearly defined
  • Time axis formatting (date range, granularity)
  • Y-axis scaling recommendations (log scale for long-term charts)

Chart Configuration Recommendations

Timeframe Selection:

  • Day Trading: 1-minute to 1-hour candles, last 5-10 days
  • Swing Trading: 4-hour to daily candles, last 3-6 months
  • Position Trading: Daily to weekly candles, last 1-3 years

Indicator Combinations:

  • Trend Following: SMA 50/200 + MACD + ADX
  • Mean Reversion: Bollinger Bands + RSI + Stochastic
  • Momentum: RSI + MACD + Volume
  • Breakout: Bollinger Bands + Volume + Pattern Recognition

Multi-Timeframe Analysis (MTF)

Framework

Analyze 3 timeframes simultaneously — higher timeframe establishes the trend, primary timeframe identifies the setup, lower timeframe refines the entry. Trade only in the direction of the higher timeframe trend.

Trading Setups by Approach

ApproachHigher TF (Trend)Primary TF (Setup)Lower TF (Entry)
Day TradingWeeklyDaily1-Hour
Swing TradingMonthlyWeeklyDaily
Position TradingQuarterlyMonthlyWeekly

MTF Alignment Scoring

Alignment ScoreInterpretation
3/3 BullishStrong long setup — all timeframes confirm uptrend
3/3 BearishStrong short setup — all timeframes confirm downtrend
2/3 AlignedModerate setup — primary direction clear, minor conflict
1/3 AlignedWeak/conflicting — avoid new positions, wait for clarity
0/3 AlignedContradiction across all timeframes — no trade

MTF Confluence Table

TimeframeTrend DirectionKey SignalSupportResistanceAlignment
Higher TF
Primary TF
Lower TF
OverallScore: X/3

MTF Integration with --chart Flag

When --chart is used, MTF analysis adds the following data tables for report generation:

MTF Summary Table:
Timeframe   Candle   MA Trend    RSI     MACD       Volume Trend   Signal
Weekly      [data]   [Up/Down]   [val]   [Bull/Bear] [Rising/Fall]  [Bullish/Bearish/Neutral]
Daily       [data]   [Up/Down]   [val]   [Bull/Bear] [Rising/Fall]  [Bullish/Bearish/Neutral]
4-Hour      [data]   [Up/Down]   [val]   [Bull/Bear] [Rising/Fall]  [Bullish/Bearish/Neutral]
Alignment Score: X/3

Rule: Never take a trade rated 1/3 or lower on MTF alignment. Wait for 2/3 minimum before entry.


Volume Profile Analysis

Key Concepts

  • Point of Control (POC): The price level with the highest volume traded over the selected period. Acts as a magnetic price level — markets frequently return to and consolidate around the POC.
  • Value Area (VA): The price range where approximately 70% of total volume was traded. Defines the "fair value zone" as agreed upon by market participants.
  • Value Area High (VAH): Upper boundary of the Value Area. Strong resistance when price is approaching from below.
  • Value Area Low (VAL): Lower boundary of the Value Area. Strong support when price is approaching from above.
  • Low Volume Node (LVN): A price range with minimal volume. Price passes through LVNs quickly — acts as a gap zone.
  • High Volume Node (HVN): A price range with dense volume. Price is magnetically attracted and consolidates around HVNs.

Volume Profile Shapes

ShapeDescriptionImplication
Normal (Bell Curve)Symmetric distribution, clear POC in the middleBalanced, range-bound market — fair value accepted
P-ShapeHigh volume at top, thin tail at bottomBuying tail below = short-covering rally, bullish
b-ShapeHigh volume at bottom, thin tail at topSelling tail above = distribution/top, bearish
Double DistributionTwo high-volume nodes with low-volume gap betweenMarket in transition — break of LVN is directional
Thin ProfileVolume evenly spread, no clear POCTrending market — follow the trend, no anchoring

Volume Profile Trading Rules

ScenarioInterpretationAction
Price > VAH (above Value Area)Bullish — market accepted higher prices, strong trendBuy breakouts above VAH with volume
Price < VAL (below Value Area)Bearish — market rejected higher prices, breakdownSell breakdowns below VAL with volume
Price at POCHigh acceptance zone — balanced, range-bound likelyFade edges of range; buy VAL, sell VAH
Price enters LVN (thin area)Fast move — price moves quickly through low-volume gapExpect acceleration; don't fade momentum
Price at HVNMagnetic zone — consolidation, range trade likelyMean-reversion strategies appropriate

Session vs. Composite Profile

  • Session Profile: Volume profile for a single trading day or week. Identifies intraday POC, VAH, VAL.
  • Composite Profile: Aggregated over weeks, months, or custom date range. Identifies major structural levels.
  • Integration rule: Composite HVNs = major structural support/resistance levels. LVNs = breakout acceleration zones. Session POC = short-term magnet.

Advanced Indicators: Ichimoku Cloud

Five Components

ComponentCalculationDefault PeriodPurpose
Tenkan-sen(9-period High + 9-period Low) / 29 periodsShort-term trend / momentum line
Kijun-sen(26-period High + 26-period Low) / 226 periodsMedium-term trend / baseline
Senkou Span A(Tenkan-sen + Kijun-sen) / 2, plotted 26 aheadLeadingFast cloud boundary
Senkou Span B(52-period High + 52-period Low) / 2, plotted 26 ahead52 periodsSlow cloud boundary
Chikou SpanCurrent close plotted 26 periods backLaggingConfirms trend vs. historical price

The shaded area between Senkou Span A and Senkou Span B is the Kumo (Cloud).

Bullish Ichimoku Signals

SignalConditionStrength
Price above CloudPrice above both Span A and Span BBase condition
Green CloudSpan A > Span B (cloud is green/bullish)Trend support
TK Cross (Bullish)Tenkan-sen crosses above Kijun-senMedium signal
TK Cross above CloudBullish TK cross while price is above the cloudStrong signal
Chikou above priceChikou Span is above historical price 26 periods agoConfirms uptrend
All 5 elements alignedAll conditions bullish simultaneouslyStrongest signal

Bearish Ichimoku Signals

SignalConditionStrength
Price below CloudPrice below both Span A and Span BBase condition
Red CloudSpan B > Span A (cloud is red/bearish)Trend resistance
TK Cross (Bearish)Tenkan-sen crosses below Kijun-senMedium signal
TK Cross below CloudBearish TK cross while price is below the cloudStrong signal
Chikou below priceChikou Span is below historical price 26 periods agoConfirms downtrend
All 5 elements alignedAll conditions bearish simultaneouslyStrongest signal

Kumo Twist

  • A Kumo Twist occurs when Senkou Span A and Span B cross — changing the cloud color from green to red (bearish twist) or red to green (bullish twist).
  • The cloud is plotted 26 periods ahead, so a Kumo Twist is an early warning signal visible in advance.
  • Thin cloud: Weak support/resistance — price can break through easily.
  • Thick cloud: Strong support/resistance — price will struggle to penetrate.

Ichimoku Signal Strength Matrix

# Bullish Conditions MetSignal GradeSuggested Bias
5/5Maximum BullishStrong long bias
4/5Strong BullishLong bias
3/5Moderate BullishLean long
2/5Mixed / TransitioningNeutral
1/5 or 0/5BearishShort bias

Options Flow Integration

Put/Call Ratio Interpretation

The Put/Call ratio measures options market sentiment — the ratio of put volume to call volume on a given day.

P/C RatioInterpretation
< 0.6Extreme bullishness / complacency — contrarian warning (overbought)
0.6–0.7Bullish — call buyers dominating, positive sentiment
0.7–1.0Neutral — balanced activity
1.0–1.3Bearish fear — put buyers dominating, defensive hedging
> 1.3Extreme fear — potential contrarian buy signal (maximum bearishness)

CBOE Total P/C vs. Equity-Only P/C: Equity-only P/C more directly reflects retail/institutional directional bets (excludes index hedging).

Unusual Options Activity (UOA) Signals

  • Large block trades: Single orders of 500+ contracts, especially OTM (out-of-the-money)
  • OTM sweeps: Aggressive market orders sweeping multiple exchanges at ask — directional conviction
  • Dark pool prints: Large, off-exchange block prints can signal institutional accumulation/distribution
  • Short-dated OTM calls on takeover candidates: One of the most reliable pre-announcement signals
  • Key UOA filters: Volume > 5x open interest, expiry within 30 days, OTM strike, market order execution

Implied Volatility (IV) vs. Historical Volatility (HV)

ConditionDescriptionOptions Strategy Implication
IV >> HVOptions are expensive — market pricing elevated movesPremium selling opportunity (write options)
IV ≈ HVOptions fairly pricedNo directional vol edge
IV << HVOptions are cheap — market underpricing movesBuying options (long vol opportunity)
  • IV Rank (IVR): Current IV relative to its 52-week range. IVR > 50 = elevated; IVR < 25 = depressed.
  • IV Percentile: % of days in the past year where IV was lower than current. More robust than IV Rank.

Max Pain Theory

  • Max Pain = The strike price at which the total value of all outstanding options (puts and calls) would expire worthless — the price that causes maximum loss for option buyers.
  • Near expiration, market prices have a tendency to gravitate toward the max pain level (due to dealer hedging flows).
  • Most useful in the final 5–7 days before monthly options expiration (OPEX).
  • Application: If stock is significantly above max pain near OPEX, short-term mean reversion is possible.

Gamma Exposure (GEX) and Spot Price Impact

  • Gamma measures how much a dealer's delta changes per $1 move in the underlying.
  • Positive GEX (Dealer Long Gamma): Dealers buy into dips and sell into rallies — suppresses volatility, market stays range-bound.
  • Negative GEX (Dealer Short Gamma): Dealers sell into dips and buy into rallies — amplifies moves, increases volatility.
  • GEX Flip Level: The price at which market transitions from positive to negative gamma — crossing this level can accelerate directional moves.
  • Practical rule: Low realized volatility + tightening range = likely high positive GEX environment. Breakouts become more powerful when negative GEX zone is entered.

Note: This section provides interpretive framework. Actual real-time options flow data, IV readings, and GEX levels must be sourced from the user's options data provider (e.g., Market Chameleon, Unusual Whales, SpotGamma, Thinkorswim).


Output

Provide technical analysis report with:

  • Current trend and momentum assessment
  • Key chart patterns identified
  • Support and resistance levels
  • Technical indicator signals
  • Volume and volatility analysis
  • Entry/exit points and price targets
  • Stop-loss recommendations
  • Risk/reward ratio for trade setups
  • Time horizon (short-term, swing, position trading)
  • MTF alignment score and summary table
  • Ichimoku Cloud status (price above/below/inside cloud)
  • Volume Profile POC, VAH, VAL levels

Enhanced Output (with --chart flag)

  • All standard output sections above
  • Chart data tables for price, volume, and indicators
  • MTF confluence table with signal direction per timeframe
  • ASCII charts for terminal display
  • Pattern recognition overlays with annotations
  • Support/resistance level visualizations
  • Volume Profile data (POC, VA, LVN/HVN zones)
  • Technical indicator panels (RSI, MACD, Bollinger Bands, Ichimoku)
  • Chart specifications for HTML report generation
  • Color-coded signals (bullish/bearish/neutral)

Keep analysis objective and data-driven, avoiding emotional bias.

Thesis Invalidation

After delivering the analysis signal, specify what would reverse it:

If signal is BULLISH — thesis breaks if:

  • Price closes below the MA200 / key support level identified in this analysis on above-average volume
  • MA30 crosses below MA60 (death cross) AND RSI drops below 40
  • Macro regime shift: Fed pivots hawkish unexpectedly, recession probability >60%

If signal is BEARISH — thesis breaks if:

  • Price closes above key resistance / MA200 level with volume confirmation
  • price reclaims MA200 AND MA30 turns up AND volume confirms
  • Fundamental improvement: surprise earnings beat >20% with guidance raise

Re-run this analysis when:

  • Next earnings release
  • Price moves ±15% from current level
  • 60 days have elapsed
  • Material news event (acquisition, leadership change, regulatory decision)

Standard Signal Output

All analysis concludes with this standardized block:

In the output, the box below is followed immediately by this machine-readable footer — the same values as the box, filled in (no placeholders, no comments), so tools can read the result without parsing the box:

{"investskill": "1", "skill": "technical-analysis", "ticker": "<TICKER, or null for a portfolio / market-wide view>", "as_of": "<YYYY-MM-DD from Data & Sources>", "signal": "<BULLISH | NEUTRAL | BEARISH>", "confidence": "<HIGH | MEDIUM | LOW>", "horizon": "<SHORT-TERM | MEDIUM-TERM | LONG-TERM>", "score": 0.0, "action": "<BUY | HOLD | SELL>", "conviction": "<STRONG | MODERATE | WEAK>", "retrieval": "<pasted by user | web/tool retrieval | model memory>", "sources": ["<each primary source, dated>"]}
╔══════════════════════════════════════════════╗
║              INVESTMENT SIGNAL               ║
╠══════════════════════════════════════════════╣
║ Signal:      BULLISH / NEUTRAL / BEARISH     ║
║ Confidence:  HIGH / MEDIUM / LOW             ║
║ Horizon:     SHORT / MEDIUM / LONG-TERM      ║
║ Score:       X.X / 10                        ║
╠══════════════════════════════════════════════╣
║ Action:      BUY / HOLD / SELL               ║
║ Conviction:  STRONG / MODERATE / WEAK        ║
╚══════════════════════════════════════════════╝

Score Guide: 8.0–10.0 Strongly Bullish | 6.0–7.9 Moderately Bullish | 4.0–5.9 Neutral | 2.0–3.9 Moderately Bearish | 0.0–1.9 Strongly Bearish Confidence: HIGH (strong data, clear signals) | MEDIUM (mixed signals) | LOW (limited data, conflicting signals) Horizon: SHORT-TERM (1 week–3 months) | MEDIUM-TERM (3 months–1 year) | LONG-TERM (1+ years)

Disclaimer: Educational analysis only. Not financial advice.

Thesis Tracker — Write It Down, Then Check It

⚠️ Data Verification — Do This Before Any Analysis

Before running any analysis, always retrieve the latest market data for the ticker:

  1. Fetch current price — use web search or ask the user for the live price, 52-week range, and market cap. Never assume a price from training data.
  2. Confirm key figures — the current value of every KPI the thesis names (revenue growth, margin, FCF, net debt, subscriber count, …). A thesis check against stale KPIs is worse than no check: it produces false confidence.
  3. State your data source — fill in the Data & Sources header (next section) so the origin, as-of date, retrieval path, and confidence of every figure are explicit at the top of the output.
  4. Flag stale data explicitly — if live data is unavailable, display this warning before proceeding:

⚠️ Live data unavailable. The following analysis uses training-data estimates which may be significantly out of date. Verify all prices and metrics before making any decisions.

Never silently substitute training-data estimates for current prices. When in doubt, ask the user to paste the latest quote.


📋 Data & Sources Header — Open Every Output With It

The first thing in the output is this provenance block, filled in — never left as placeholders. It is the standard documented on the Data & Accuracy page and the first thing result-validator looks for:

Data & Sources
  As of:      <date the figures represent, e.g. 2026-06-30>
  Source:     <primary docs — SEC EDGAR 10-K/10-Q, company IR, FRED, exchange data …>
  Retrieval:  <pasted by user | web/tool retrieval | model memory>
  Confidence: <HIGH | MEDIUM | LOW>
  • Retrieval: model memory must be paired with Confidence: LOW — memory is a placeholder until confirmed against a primary source.
  • Mixed sources: list each with its own as-of date rather than blending them.
  • Sources disagree on the same figure (a closing price, a deal term): prefer the primary source (SEC filing, company release, exchange data) over an aggregator, then the most recent dated one. Show both values with their sources and dates, say which you used and why, and never average them.
  • Data the user pasted is reported as pasted by user; do not upgrade its confidence beyond what the user's own source supports.

🌐 Output Language — --lang zh-TW

English is the default. With --lang zh-TW — or when the user writes in Traditional Chinese and asks for a Chinese answer — write the whole analysis in Traditional Chinese (繁體中文, Taiwan usage), keeping it machine-checkable:

  • Translate every heading, narrative paragraph, table header and bullet. Give each metric's English name in brackets the first time it appears — 自由現金流(Free Cash Flow, FCF) — so the reader can match it to the glossary and to English analyses.
  • Keep in English tickers, figures and units, SEC form and item names (10-K Item 7, Form 4), the four Data & Sources field labels (As of, Source, Retrieval, Confidence — the Retrieval value may be written in Chinese, e.g. 使用者提供), and the Investment Signal box and JSON footer exactly as specified — field names and the BULLISH / NEUTRAL / BEARISH, HIGH / MEDIUM / LOW, BUY / HOLD / SELL, STRONG / MODERATE / WEAK values. Tools parse them.
  • Title the invalidation section 論點失效條件(Thesis Invalidation), and close with the disclaimer in Chinese: 免責聲明: 僅供教育用途,非投資建議。
  • Use Taiwan financial terms: 殖利率、本益比、營業利益率、自由現金流、聯準會、選擇權、買權/賣權、除息日、預扣稅.

Overview

Every other framework in the catalog answers a question today: is it cheap, is the moat real, is the chart constructive, what does the bear say. None of them remembers the answer. Three months later the investor is left with a position, a vague feeling, and no record of what they expected to happen — which is exactly the state in which people average down on a broken story or sell a working one on a bad week.

This skill turns an analysis into a contract with your future self. It produces a single saved file per ticker that states, in falsifiable terms, why the position exists, which numbers have to keep being true, what would prove the idea wrong, and when the next check is due. Then, on --update, it re-reads that file against current data and hands back one word — INTACT, WEAKENED, or BROKEN (or INSUFFICIENT EVIDENCE when the data to decide is not there) — together with the specific line that changed.

It fills the gap between Playbook steps 1 (write the thesis), 9 (monitor), and 10 (know when to sell) in the Learning track, which today have no skill behind them.

What it is not. It is not a valuation, a screen, or a timing tool. It imports its KPIs and triggers from the other skills (stock-eval for quality metrics, stock-valuation for the fair-value range, bear-case for the thesis-killers, catalyst-calendar for dates, position-ladder for the share-count ceiling) and holds them to account. Run the analysis skills first; run this one to write down what they concluded.


1. Modes

ModeInvocationWhat it does
Openthesis-tracker NVDA (+ pasted analyses, or none)Interview the user, draft the thesis file, save it to output/thesis/NVDA.md
Updatethesis-tracker NVDA --update (+ current data or a fresh analysis)Re-read the saved file, refresh every KPI, re-test every trigger, return INTACT / WEAKENED / BROKEN (or INSUFFICIENT EVIDENCE), append to the decision log
Reviewthesis-tracker --review (no ticker)Read every file in output/thesis/, list status and next-check date per ticker, flag anything overdue or sitting at INSUFFICIENT EVIDENCE
Closethesis-tracker NVDA --close "sold — trigger 2 fired"Mark the thesis closed with the exit reason and final P&L; keep the file for trade-postmortem

If the environment cannot write files, output the complete thesis file as a fenced block and tell the user to save it at the stated path — the contract matters, not the storage.


2. Inputs

InputRequiredDefault if unstated
Ticker✅ for Open, Update, Close · not used by Review (it reads every file)—
The thesis in the user's own words✅ for Openask — never invent a thesis for the user
Entry price and date, position sizerecommendedrecord as "not stated"
Prior analyses (stock-eval, bear-case, stock-valuation, …)optionalderive KPIs from the thesis text
Horizonoptional12 months
Existing thesis file✅ for Update / Close—

The user writes the thesis; the skill sharpens it. If the stated thesis is not falsifiable ("great company, will go up"), push back with the three questions in Phase 1 until it is. A thesis that cannot be wrong cannot be tracked.


3. Framework

Phase 1 — Sharpen the thesis (Open)

Reduce the thesis to one paragraph that answers, in order:

  1. What has to happen to the business for this to work (the mechanism — not "the stock goes up")
  2. Why the market is wrong today — what is mispriced, and why the user sees it and the marginal buyer does not
  3. Why now — the catalyst, the regime, or the price that makes this a decision today instead of in a year

Reject any paragraph that only restates the company description. Test: could a reasonable bear read this and know exactly what to disagree with?

Phase 2 — Pick the KPIs (3–5, no more)

Each KPI is a number the thesis depends on, with a threshold below or above which the thesis is in trouble, and a source that can be checked next quarter.

KPIThresholdWhy it matters to this thesisSourceCurrentDate
e.g. Data-center revenue growth (YoY)≥ 40%The multiple assumes hyperscaler capex keeps compounding10-Q segment table112%2026-05-28
e.g. Gross margin≥ 70%Pricing power is the moat claim; margin is its footprint10-Q75.5%2026-05-28
e.g. Net cash> $0Rules out balance-sheet stress in a downturn10-Q+$38B2026-05-28

Rules:

  • Prefer KPIs from the financial statements over sentiment or price-based ones. Price is the output, not an input to the thesis.
  • One valuation KPI is allowed (e.g. forward P/E ≤ 35 or price within stock-valuation's fair-value range) so that "the thesis is right but fully priced" is a detectable state.
  • If a KPI has no threshold, it is a curiosity, not a KPI. Remove it.
  • Import thresholds from the analyses that produced them when available (stock-eval quality gates, dividend-analysis payout ceiling, short-interest squeeze levels).

Phase 3 — Write the invalidation triggers

Triggers are events, not feelings. Each one names the observation and the action it obliges:

Triggers   | if <observable event or KPI breach>        → <action: review / trim / exit>

Draw them from three places, in this order of authority:

  1. The Thesis Invalidation sections of every analysis skill already run on the ticker — copy them in verbatim, they were written for exactly this
  2. The bear-case Thesis-Killers — the bear's best arguments become the bull's triggers
  3. The user's own "I would sell if…" — written down while calm

A KPI breach is a review trigger by default; two simultaneous breaches, or any breach of the thesis's core mechanism (Phase 1, item 1), is an exit trigger. Price alone is never an exit trigger unless the user is explicitly running a stop-loss discipline — write that down too, so it is a rule and not a reaction.

Phase 4 — Catalysts and the next check

List the dated events that will produce new evidence for or against the thesis (earnings, product cycles, regulatory decisions, lock-up expiries, FOMC when the thesis is rate-sensitive). Import from catalyst-calendar when available. The next-check date is the earliest of: the next catalyst, the next earnings release, or 90 days.

Phase 5 — The pre-mortem

One paragraph, written in the past tense, dated twelve months out:

"It is <date + 12 months>. The position lost 40%. Looking back, the reason was…"

This forces the most probable failure path into the open while it can still be wired into a trigger. If the pre-mortem names a cause that no trigger would catch, add the trigger.

Phase 6 — Update: re-test everything (--update)

On every update:

  1. Refresh the Current / Date columns of every KPI from the newest data the user supplies or the assistant can retrieve. Never carry a stale value forward silently — mark it stale with its date.
  2. Re-test every trigger: fired / not fired / cannot assess (say why).
  3. Re-read the thesis paragraph against what actually happened. Did the mechanism play out, stall, or go the other way?
  4. Return exactly one status. Apply the rules top-down; the first that matches wins:
StatusRule
BROKENTwo or more KPIs breached on current data, or any exit-trigger fired, or the core mechanism has been falsified (the thing that had to happen did not, and the reason is structural)
WEAKENEDOne KPI breached or one review-trigger fired or the mechanism is behind schedule — the thesis is still possible but the evidence has moved against it
INSUFFICIENT EVIDENCEFewer than half the KPIs could be refreshed with data dated after the last check, and no trigger could be assessed. Nothing is known to have broken — but nothing has been confirmed either. Name the KPIs that need data, and do not carry the previous status forward
INTACTEvery KPI is within threshold on data dated after the last check; no trigger fired; mechanism on track. A stale KPI cannot support INTACT — at best it leaves the status at INSUFFICIENT EVIDENCE

An INSUFFICIENT EVIDENCE status does not change the file's last confirmed status line; it is recorded in the decision log (checked — insufficient evidence — <KPIs missing>) and the next-check date is moved to when the data will exist (usually the next filing).

  1. Name the specific line that changed — the KPI, its old value, its new value, and the threshold. A status without a line is an opinion.
  2. Append a decision log row. The log records what the user did and why, priced and dated, and names the skill that informed it. It is the raw material for trade-postmortem.

Status is not a signal. WEAKENED does not mean sell; it means the next check moves up and the position should not be added to. BROKEN means the reason for owning it is gone — what to do about that is a position-ladder / tax question, but adding is off the table. INSUFFICIENT EVIDENCE means the honest answer is "I don't know yet" — treat it like WEAKENED for position management (no adding) until the data arrives.

Phase 7 — Thesis Health Score (0–10)

The headline number, used for the signal block. It measures how well the thesis is holding up against its own tests, not how attractive the stock is.

ComponentWeight10 looks like0 looks like
KPI adherence40%All KPIs inside threshold with marginMajority breached
Trigger status30%None firedExit trigger fired
Mechanism progress20%Phase 1 "what has to happen" is visibly happeningFalsified
Evidence freshness10%Every KPI dated within the last quarterKPIs stale or unverifiable

Map — the same bands as the standard Score Guide: ≥ 6.0 → INTACT · 4.0–5.9 → WEAKENED · < 4.0 → BROKEN. INSUFFICIENT EVIDENCE is not scored from the table above — report the score as n/a in the file and use 5.0 in the signal block (NEUTRAL, Confidence LOW) so it stays comparable across skills. If the rule table in Phase 6 and the score disagree, the rule table wins and the score is adjusted to match — the score summarizes the rules, it does not override them.


4. The Saved File — Contract

output/thesis/<TICKER>.md. Keep the section order and the pipe tables exactly; --update and --review parse them.

# Thesis · <TICKER> · opened YYYY-MM-DD · status INTACT | WEAKENED | BROKEN | INSUFFICIENT EVIDENCE | CLOSED
Data & Sources · as of YYYY-MM-DD · <sources> · <retrieval> · <confidence>

## Thesis
<one paragraph: what has to happen · why the market is wrong · why now>
Horizon: <n> months   Entry: $<price> on YYYY-MM-DD   Size: <shares or % of portfolio>   Ceiling: <from position-ladder, if any>

## KPIs
| KPI | Threshold | Source | Current | Date | Status |
|-----|-----------|--------|---------|------|--------|
| … | … | … | … | … | ✓ / ✗ / stale |

## Triggers
| If | Then | Source | Status |
|----|------|--------|--------|
| … | review / trim / exit | bear-case · stock-eval · user | not fired / FIRED YYYY-MM-DD |

## Catalysts
| Date | Event | Expected impact | Evidence for / against |
|------|-------|-----------------|------------------------|

## Pre-mortem
"It is YYYY-MM-DD. This lost 40% because …"

## Next check
YYYY-MM-DD — <reason: earnings / catalyst / 90-day>

## Decision log
| Date | Action | Price | Reason | Informed by |
|------|--------|-------|--------|-------------|
| YYYY-MM-DD | opened | $… | … | stock-eval, bear-case |

5. Output Format

Open mode

  1. Data & Sources header
  2. The sharpened thesis paragraph, with a one-line note on what was changed from the user's draft and why
  3. The KPI table, the trigger table, the catalyst list, the pre-mortem, the next-check date
  4. The complete saved file in a fenced block (and confirmation of the path it was written to)
  5. Thesis Invalidation and the Investment Signal block

Update mode

  1. Data & Sources header
  2. Status line first: NVDA · WEAKENED · 2026-08-29 — Gross margin 68.1% (was 75.5%) breached the ≥ 70% threshold
  3. KPI table with old → new values and ✓/✗; trigger table with fired/not fired
  4. What the change means for the Phase 1 mechanism, in three sentences
  5. The updated file (fenced) with the new decision-log row
  6. Thesis Invalidation and the Investment Signal block

Review mode — one table: ticker · status · score · next check · overdue? · last decision.


Example

User: thesis-tracker NVDA — I bought at $118 on 2026-05-30. Thesis: data-center demand
      compounds for at least two more years and the market is still pricing a one-off cycle.
      Here are my stock-eval and bear-case outputs. [pasted]

The assistant sharpens the paragraph (adds "why now": hyperscaler FY27 capex guides due in
the next two prints), sets four KPIs (DC revenue growth ≥ 40%, gross margin ≥ 70%, net cash > 0,
forward P/E ≤ 35), imports three triggers from the bear-case Thesis-Killers (a top-3 customer
announcing in-house silicon at scale; export-control expansion to a second major market;
inventory days > 120), lists the next two earnings dates and one export-policy decision as
catalysts, writes the pre-mortem ("…because the 2027 capex guides came in flat and the
multiple compressed before revenue did"), and saves output/thesis/NVDA.md with status INTACT
and a next check on the August print.

User: thesis-tracker NVDA --update [pastes the August 10-Q highlights]

The assistant returns: NVDA · WEAKENED · 2026-08-29 — Gross margin 68.1% (was 75.5%) breached
the ≥ 70% threshold; the other three KPIs hold; no trigger fired. Notes that the pricing-power
leg of the thesis is the one under pressure, moves the next check to the October analyst day,
and appends the log row "2026-08-29 · hold, no add · $131 · margin breach under review".

Notes

  • Run the analysis skills first. This skill is a notebook with rules; it is only as good as the analyses it summarizes. Open with at least stock-eval and bear-case in hand.
  • Never write the thesis for the user. Sharpen, question, and structure — but the conviction has to be theirs or the file will not be believed when it says BROKEN.
  • Hold the KPI count at five. More KPIs means more noise and a guaranteed breach every quarter; the thesis becomes permanently WEAKENED and the status stops carrying information.
  • A WEAKENED status blocks adding, not holding. Pair with position-ladder, whose thesis-break gate is meant to read this status.
  • Closed files are inputs to trade-postmortem. Do not delete them.
  • File paths and the output/ convention follow full-report. Where the environment cannot write, print the file.

Thesis Invalidation

After delivering the analysis signal, specify what would reverse it:

If signal is BULLISH (thesis INTACT) — it turns WEAKENED or BROKEN if:

  • Any KPI in the file crosses its threshold at the next check (name the KPI closest to its threshold today and by how much)
  • Any trigger in the file fires — the trigger table is the invalidation list for this thesis
  • The core mechanism stalls for two consecutive checks even with KPIs technically inside threshold (the numbers hold but the story does not)

If signal is BEARISH (thesis BROKEN) — the verdict is wrong if:

  • The breached KPI was a one-off (a charge, a reclassification, a timing shift) and reverts within one quarter — say what evidence would show that
  • The fired trigger was mis-specified (it caught a symptom that turned out not to matter to the mechanism) — rewrite the trigger rather than ignore it
  • The user's original thesis was narrower than the one written down — re-open with the honest, narrower version

Re-run this analysis when:

  • The next-check date in the file arrives
  • Any catalyst in the file occurs
  • Price moves ±20% from the entry or the last check (a check, not a reaction)
  • A new bear-case or stock-eval run produces triggers the file lacks
  • 90 days have elapsed since the last update

Standard Signal Output

This skill measures thesis health, not stock direction, so state the mapping: Signal: BULLISH means the thesis is INTACT (the reason for owning it holds); NEUTRAL means WEAKENED (hold, do not add, check sooner) — or INSUFFICIENT EVIDENCE, in which case Confidence: LOW and Action: HOLD are mandatory; BEARISH means BROKEN (the reason for owning it is gone). Action: BUY here means "the thesis supports continuing to hold or ladder within the stated ceiling" — never "buy without limit." Read the direction of the stock from stock-eval or bear-case, not from this block.

All analysis concludes with this standardized block:

In the output, the box below is followed immediately by this machine-readable footer — the same values as the box, filled in (no placeholders, no comments), so tools can read the result without parsing the box:

{"investskill": "1", "skill": "thesis-tracker", "ticker": "<TICKER, or null for a portfolio / market-wide view>", "as_of": "<YYYY-MM-DD from Data & Sources>", "signal": "<BULLISH | NEUTRAL | BEARISH>", "confidence": "<HIGH | MEDIUM | LOW>", "horizon": "<SHORT-TERM | MEDIUM-TERM | LONG-TERM>", "score": 0.0, "action": "<BUY | HOLD | SELL>", "conviction": "<STRONG | MODERATE | WEAK>", "retrieval": "<pasted by user | web/tool retrieval | model memory>", "sources": ["<each primary source, dated>"]}
╔══════════════════════════════════════════════╗
║              INVESTMENT SIGNAL               ║
╠══════════════════════════════════════════════╣
║ Signal:      BULLISH / NEUTRAL / BEARISH     ║
║ Confidence:  HIGH / MEDIUM / LOW             ║
║ Horizon:     SHORT / MEDIUM / LONG-TERM      ║
║ Score:       X.X / 10                        ║
╠══════════════════════════════════════════════╣
║ Action:      BUY / HOLD / SELL               ║
║ Conviction:  STRONG / MODERATE / WEAK        ║
╚══════════════════════════════════════════════╝

Score Guide: 8.0–10.0 Strongly Bullish | 6.0–7.9 Moderately Bullish | 4.0–5.9 Neutral | 2.0–3.9 Moderately Bearish | 0.0–1.9 Strongly Bearish Confidence: HIGH (strong data, clear signals) | MEDIUM (mixed signals) | LOW (limited data, conflicting signals) Horizon: SHORT-TERM (1 week–3 months) | MEDIUM-TERM (3 months–1 year) | LONG-TERM (1+ years)

Note: The Score above is the Thesis Health Score, mapped onto the standard scale for cross-skill comparability. It rates how well the written thesis is surviving its own tests — not how attractive the stock is.

Disclaimer: Educational analysis only. Not financial advice. A thesis file is a record of the user's reasoning, not a recommendation to hold, add to, or sell any position.

Trade Post-Mortem

⚠️ Data Verification — Do This Before Any Analysis

Before running any analysis, always retrieve the latest market data for the ticker:

  1. Fetch current price — use web search or ask the user for the live price, 52-week range, and market cap. Never assume a price from training data.
  2. Confirm key figures — every fill (date, side, shares, price, fees) from the thesis file's decision log or the broker export; dividends received while held; the total-return closes of SPY and the chosen sector ETF on the exact entry and exit dates (adjusted for distributions); the stock's beta over the holding window; and the KPI values that were current on the day each decision was made. A post-mortem built on remembered fills or end-of-month benchmark prices grades the wrong trade.
  3. State your data source — fill in the Data & Sources header (next section) so the origin, as-of date, retrieval path, and confidence of every figure are explicit at the top of the output.
  4. Flag stale data explicitly — if live data is unavailable, display this warning before proceeding:

⚠️ Live data unavailable. The following analysis uses training-data estimates which may be significantly out of date. Verify all prices and metrics before making any decisions.

Never silently substitute training-data estimates for current prices. When in doubt, ask the user to paste the latest quote.


📋 Data & Sources Header — Open Every Output With It

The first thing in the output is this provenance block, filled in — never left as placeholders. It is the standard documented on the Data & Accuracy page and the first thing result-validator looks for:

Data & Sources
  As of:      <date the figures represent, e.g. 2026-06-30>
  Source:     <primary docs — SEC EDGAR 10-K/10-Q, company IR, FRED, exchange data …>
  Retrieval:  <pasted by user | web/tool retrieval | model memory>
  Confidence: <HIGH | MEDIUM | LOW>
  • Retrieval: model memory must be paired with Confidence: LOW — memory is a placeholder until confirmed against a primary source.
  • Mixed sources: list each with its own as-of date rather than blending them.
  • Sources disagree on the same figure (a closing price, a deal term): prefer the primary source (SEC filing, company release, exchange data) over an aggregator, then the most recent dated one. Show both values with their sources and dates, say which you used and why, and never average them.
  • Data the user pasted is reported as pasted by user; do not upgrade its confidence beyond what the user's own source supports.

🌐 Output Language — --lang zh-TW

English is the default. With --lang zh-TW — or when the user writes in Traditional Chinese and asks for a Chinese answer — write the whole analysis in Traditional Chinese (繁體中文, Taiwan usage), keeping it machine-checkable:

  • Translate every heading, narrative paragraph, table header and bullet. Give each metric's English name in brackets the first time it appears — 自由現金流(Free Cash Flow, FCF) — so the reader can match it to the glossary and to English analyses.
  • Keep in English tickers, figures and units, SEC form and item names (10-K Item 7, Form 4), the four Data & Sources field labels (As of, Source, Retrieval, Confidence — the Retrieval value may be written in Chinese, e.g. 使用者提供), and the Investment Signal box and JSON footer exactly as specified — field names and the BULLISH / NEUTRAL / BEARISH, HIGH / MEDIUM / LOW, BUY / HOLD / SELL, STRONG / MODERATE / WEAK values. Tools parse them.
  • Title the invalidation section 論點失效條件(Thesis Invalidation), and close with the disclaimer in Chinese: 免責聲明: 僅供教育用途,非投資建議。
  • Use Taiwan financial terms: 殖利率、本益比、營業利益率、自由現金流、聯準會、選擇權、買權/賣權、除息日、預扣稅.

Overview

Every other skill in the catalog looks forward. thesis-tracker writes the contract before the trade and re-checks it while the position is open; position-ladder plans the entries and trims; tax-lens prices the exit. Once the position is closed, nothing looks back — and looking back is the only way an investor learns whether their process works, as distinct from whether they were lucky this time.

This skill grades the decision separately from the result. It takes a closed trade (or a year of them), reconstructs what the investor knew and planned at each decision point, and answers five questions in order: Which quadrant of the process-vs-outcome matrix is this? Did I do what my plan said? Which biases showed up, and where is the evidence? Were the pre-trade analyses right, and for the right reasons? Where did the return actually come from? It ends with one sentence for the journal — the lesson — and a Process Score 0–10.

It reuses the vocabulary of Learning lesson 13, Psychology & Process (site/content/LEARNING-PSYCHOLOGY.md): the four quadrants (deserved success, bad luck, dumb luck, deserved loss), the eight biases and their antidotes, the journal template, and the sell discipline of rules vs. reasons.

What it is not. It is not a re-analysis of the stock with hindsight, and it is not a tax calculation (tax-lens does that). The signal block at the end is the only forward-looking part: the view on re-entering the ticker today, informed by what the post-mortem learned. Pairs with thesis-tracker (its closed files are this skill's primary input), position-ladder (the plan the sizing is graded against), tax-lens (after-tax result and wash-sale check on any re-entry), and learning-coach (to turn the lesson into a drill).


1. Modes and Inputs

ModeInvocationWhat it does
Single tradetrade-postmortem ZEPHReads output/thesis/ZEPH.md (status CLOSED) or a pasted trade log, grades one round trip
Yeartrade-postmortem --year 2026Reads every CLOSED file in output/thesis/ (and/or a broker CSV) with a close date in the year; per-trade grades plus aggregate statistics
InputRequiredDefault if unstated
Ticker (single mode) or year (year mode)✅—
Closed thesis file output/thesis/<TICKER>.md or a pasted trade log / broker CSV✅ one of the two—
BenchmarksoptionalSPY plus the SPDR sector ETF matching the stock's GICS sector (XLK, XLV, XLF, …); state the choice
Prior analysis outputs (stock-eval, bear-case, stock-valuation, earnings-preview, …) dated before the decisionsrecommendeduse the "Informed by" column of the decision log; if none survive, say the pre-trade comparison is limited
position-ladder plan (rungs, ceiling)optionalthe Ceiling: field in the thesis file
Account type and tax bracketoptionalpre-tax only, with a one-line pointer to tax-lens
Ticker to evaluate for re-entry (year mode)optionalthe largest-loss trade of the year — the most likely candidate for a revenge trade

Reading a closed thesis-tracker file

The file follows the thesis-tracker saved-file contract exactly; read it section by section:

File sectionWhat the post-mortem takes from it
# Thesis · <TICKER> · opened YYYY-MM-DD · status CLOSEDOpen date; confirm status is CLOSED (if not, stop — run thesis-tracker <TICKER> --close first, or treat as a pasted log and say so)
Data & Sources lineThe as-of date of the original analysis — the information set the entry decision is graded against
## Thesis paragraph and Horizon / Entry / Size / Ceiling lineThe mechanism, the planned horizon, the planned entry, size, and share-count ceiling
## KPIs tableThe thresholds and the last recorded values — did a KPI breach precede the exit?
## Triggers tableWhich triggers fired and when (FIRED YYYY-MM-DD), and what action each obliged (review / trim / exit)
## CatalystsWhich expected events happened, and what evidence they produced
## Pre-mortemDid the actual failure (or success) path match the one written in advance?
## Decision logEvery action, price, reason, and the skill that informed it — the raw material for plan adherence and bias detection. The final closed row carries the exit reason and P&L written by --close

Reading a pasted trade log or broker CSV

Accept any export with at least date, side (buy/sell), ticker, quantity, price; use fees, dividends, and lot id when present. Match sells to buys by specific lot ID if given, else FIFO, and say which. Without a written plan there is nothing to adhere to: reconstruct the implied plan from the user's own words (ask: "what were you expecting, and what would have made you sell?"), label it reconstructed, and cap the plan-adherence component at 5 / 10 — a plan written after the fact is shaped by the outcome.


2. Framework

Phase 1 — Reconstruct the timeline (the hindsight firewall)

Build one table of every decision, with only the information available on that date:

DateActionShares · pricePosition afterThesis status thenKPI / trigger state thenReason given thenInformed by
YYYY-MM-DDopen / add / trim / close / reviewshares · % of portfolioINTACT / WEAKENED / BROKENe.g. GM 58.2% vs ≥ 55% ✓from the decision logskill + run date

The hindsight rule: a decision is graded on what was knowable when it was made. "The CEO resigned three months later" is not evidence the entry was bad; "the 10-Q that showed the margin breach was out a week before the add" is evidence the add was bad. Mark any fact the user only learned later as (later) and keep it out of the process grade.

Phase 2 — The outcome, honestly measured

MeasureHowValue
Gross P&L ($)proceeds − cost basis, all lots
Dividends received ($)sum while held (ex-dates inside the window)
Costs ($)commissions + fees + estimated spread (½ bid-ask × shares per fill) + FX if any
Total return after costs(gross P&L + dividends − costs) ÷ total capital deployed%
Money-weighted return (IRR)when there were multiple entries/exits; it charges the adds and trims for their timing% annualized
Holding periodfirst buy → last sell; and the weighted average by lotdays
SPY total return, same windowfirst-entry date → exit date, distributions reinvested%
Sector ETF total return, same windowas above%
Excess vs. benchmarktotal return − benchmark return (use the sector ETF when the thesis was about the company within its industry, SPY when it was a market or macro call; state which)pp
Tax noteshort- vs. long-term (> 1 year), wash-sale risk if re-bought within 30 days — the numbers go to tax-lensone line

Outcome grade (column of the matrix): good outcome = excess total return after costs vs. the chosen benchmark ≥ 0; bad outcome = < 0. Report the absolute return beside it, and when the two disagree say so in one line ("made 6%, but the sector ETF made 14% — the capital would have done better in the index").

Phase 3 — Return attribution: where the money came from

Decompose the stock's own price-plus-dividend return over the window (first entry → final exit, as if one lot) into three additive layers:

LayerFormulaReads as
MarketR_SPYWhat simply being in stocks earned
SectorR_sector − R_SPYWhat being in this industry added or cost
Stock-specificR_stock − R_sectorWhat this company did on its own — the only layer the thesis claimed to know
(check)the three sum to R_stock

When the stock's beta over the window is far from 1 (below 0.7 or above 1.3), add the beta-adjusted view: expected = β × R_SPY; alpha = R_stock − β × R_SPY. A high-beta winner in a rising market may have zero alpha.

Then a fourth, separate number — timing and execution:

Plan return     = return had every fill followed the written plan (planned rungs, ceiling,
                  and an exit on the day the first exit-trigger FIRED)
Timing effect   = actual total return − plan return

A positive timing effect means the deviations helped this time (that is still a process question — see Phase 5); a negative one is the dollar cost of not following the plan. If no written plan exists, report timing against the simplest alternative — the whole position bought on the first-entry date and sold on the exit date — and label it.

Phase 4 — Plan adherence

Grade each element against what the thesis file (or reconstructed plan) said before the trade:

ElementPlan saidWhat happenedAdhered?Cost of the deviation
Entryprice / rung / condition✓ / partial / ✗$ or pp vs. plan
Sizinginitial size, % of portfolio, cap
Addsrungs, and "no add while WEAKENED / BROKEN"
Ceilingshare-count ceiling from position-laddermax shares actually held
Stop / exit triggerwhich trigger, action obligeddate it fired → date acted on (the lag)
Profit exit / trimtarget or valuation range
Reviewsnext-check dates in the filedone on time / late / skipped

Score each row 10 (followed), 5 (partially, or followed late), 0 (ignored); the adherence sub-score is their average over the rows the plan actually covered. Two rules: an exit-trigger lag of more than 10 trading days counts as partial, more than 30 as ignored. And a deviation that made money is still a deviation — score it as one, and record its sign in the cost column so year mode can test whether the investor's overrides add value on average.

Phase 5 — Biases detected

For each bias from lesson 13, look for its evidence pattern in the decision log. A bias is confirmed only when the log shows the pattern; it is suspected when the pattern is partly there or the user's own words suggest it; otherwise not seen. Never diagnose a bias from the outcome alone.

BiasEvidence pattern in the logVerdict
Anchoring on purchase priceReasons that reference cost basis: "sell when back to even", "waiting for $X" where $X is the entry; exit dates cluster near the cost basisconfirmed / suspected / not seen
Disposition effectWinner sold on a gain with thesis still INTACT and no rule fired; loser held past a fired exit trigger; holding period of losers ≫ winners (year mode test, Phase 7)
FOMO / herdingEntry after a > 20% run in the prior 20 sessions; thesis paragraph dated the same day as (or after) the entry; reasons citing social media, "everyone", or the chart alone; the 24-hour cooling-off skipped
Confirmation biasNo bear-case run before the entry or at reviews; KPIs re-defined or thresholds loosened after a breach; reviews that cite only bullish sources
Sunk costAdds made while the thesis was WEAKENED or BROKEN, or above the ceiling, with reasons like "lower my average"
OverconfidenceSize above the stated cap; cap raised mid-trade; "high conviction" used to justify size; no pre-mortem written
Loss aversionExit trigger fired and the position was held to "avoid locking in the loss"; stop moved down
Recency / extrapolationEntry or add justified by the last few quarters' trend alone; target derived by extending recent growth without a base rate

Quote the decision-log line that is the evidence. A bias without a quoted line is a guess — mark it suspected at most.

Phase 6 — What the pre-trade skills said vs. what happened

For every analysis listed in the "Informed by" column (and any others the user pastes, dated before the decision):

Skill · run dateWhat it said (signal, score, key claim)What happenedVerdict
e.g. stock-eval · 2025-02-28BULLISH 7.1 — margin expansion to 60%GM peaked at 58%, then fell to 52%wrong on the mechanism
e.g. bear-case · 2025-02-28Thesis-killer #2: competitor pricingCompetitor cut prices in Q3right — and the user ignored it

Verdicts: right for the right reason, right for the wrong reason (the call worked but the stated mechanism did not happen — treat as luck), wrong, not testable (the window was too short). Then answer: did the user follow the analysis, override it, or not run it? An override that was right is still worth recording; three overrides in a row that were right is evidence of edge, one is not.

Phase 7 — Year mode: aggregate statistics

Run Phases 1–6 per trade (abbreviated: one line each), then compute across the closed trades of the year:

StatisticFormulaHealthy reading
Trades closedcount (a round trip = one ticker from first buy to flat)—
Hit ratewinners ÷ trades (winner = total return after costs > 0)Means little alone — read with payoff
Average win / average lossmean % (and $) of winners; of losers
Payoff ratioaverage win ÷ |average loss|> 1.5 makes a 45% hit rate profitable
Expectancy per tradehit rate × avg win − (1 − hit rate) × |avg loss|> 0, and stated in % and $
Profit factorgross gains ÷ gross losses> 1.5
Largest loss vs. planthe worst trade's actual loss ÷ its planned maximum loss (stop distance × planned size)≤ 1.2×; > 2× means the stop or the ceiling did not hold
Holding period: winners vs. losersmedian days held, winners and losers separatelyLosers held ≥ 1.5× as long as winners, and average win < average loss → disposition effect flagged
Excess vs. SPY, whole yearmoney-weighted return of the closed trades − SPY over the same capital-weighted windows≥ 0
Quadrant countsnumber of trades in each of the four cellsMost trades in the top row
Bias frequencyconfirmed count per biasThe most frequent bias is next year's rule
Override valuesum of the "cost of deviation" column across all tradesNegative → follow the plan; positive over 10+ overrides → the plan needs rewriting, not ignoring
Average Process Scoremean, and its correlation with excess return

Sample-size warning: fewer than 20 closed trades makes every rate above an anecdote. Say so, report the numbers anyway, and lean on the per-trade process grades, which do not need a sample.

Phase 8 — Process Score (0–10) and the quadrant

The headline number of the post-mortem. It grades the decisions, on the information available at the time — not the P&L.

ComponentWeight10 looks like0 looks like
Pre-trade plan quality25%Falsifiable thesis, 3–5 KPIs with thresholds, exit triggers, pre-mortem, size inside cap, bear-case read — all dated before the first buyNo written plan; or written after the entry
Plan adherence (Phase 4)35%Every row followed; exit acted on within 10 trading days of the triggerCeiling breached, adds into a WEAKENED thesis, fired exit trigger ignored
Bias exposure (Phase 5)25%Start at 10; −2 per confirmed bias, −1 per suspected; floor 0Three or more confirmed biases
Review discipline15%Every next-check done on time, decision log written at the time of each action, --update run after each catalystReviews skipped; log reconstructed afterwards
Process Score = 0.25 × Plan + 0.35 × Adherence + 0.25 × Bias + 0.15 × Review

Quadrant: Process Score ≥ 6.0 → good decision row; < 6.0 → bad decision row (5.0–5.9 is reported as "borderline", but it still goes in a row). Column from Phase 2. Name the quadrant with lesson 13's label and its instruction:

Good outcomeBad outcome
Good decisionDeserved success — repeat the process; do not raise size because it workedBad luck — do not "fix" a process that was right; confirm the thesis actually broke rather than the price just moving
Bad decisionDumb luck — the most dangerous cell; write down exactly what was skippedDeserved loss — fix the process, not the stock

Phase 9 — The journal lesson

One sentence, in the first person, that names the behaviour (not the stock), the evidence, and the rule that would have prevented or repeated it:

"When the file says WEAKENED, the ladder is closed — I added 150 shares into a margin miss and it cost me $1,620 against my own plan."

Rules: no lesson about the market ("tech was weak"), no lesson that needs hindsight ("I should have known the CEO would leave"), and no more than one. If the quadrant is deserved success or bad luck, the lesson is what to keep doing. In year mode, the lesson comes from the most frequent confirmed bias or the largest override cost. Format it as a journal entry row so it can be pasted into the decision log.

Phase 10 — Re-entry view (the only forward-looking part)

The signal block answers one question: knowing what this post-mortem found, would I buy this ticker today, at today's price? It is lesson 13's anchoring antidote applied after the fact — the old cost basis and the old P&L are irrelevant.

ComponentWeight10 looks like0 looks like
Thesis today40%The original mechanism is still alive and the exit was a process error or a valuation trim; or a fresh stock-eval / thesis-tracker reading is INTACT-gradeThe mechanism was falsified structurally (Phase 6: "wrong on the mechanism")
Price vs. value today30%Current price inside or below a recent stock-valuation rangeAbove the range; or no valuation available (score 5 and cap Confidence at LOW)
Fit with the lesson30%The lesson is about execution that a rule now fixes; the user has shown edge in this kind of setupThe lesson says the user has no edge here (repeated dumb luck / deserved loss on the same setup, or the name sits outside their circle of competence)

If the user supplies no current data on the ticker, report the re-entry Score as 5.0 (NEUTRAL, Confidence LOW, Action HOLD — "no view without fresh work"). A re-entry is never a re-opening of the old file: it requires a new thesis-tracker Open with a new thesis, and a tax-lens wash-sale check when the exit was at a loss within the last 30 days.


3. Output Format

Single-trade mode

  1. Data & Sources header (fills, benchmark prices, dividends; the thesis file path and its as-of date)
  2. Verdict line first: ZEPH · Deserved loss · Process 5.6 / 10 · −20.0% after costs vs. sector ETF +4.0% · lesson below
  3. Executive summary — three sentences: what happened, which quadrant and why, what changes
  4. Timeline with the hindsight firewall (Phase 1)
  5. Outcome table (Phase 2) and attribution — three layers, beta note if needed, timing effect (Phase 3)
  6. Plan adherence table with the cost of each deviation (Phase 4)
  7. Biases detected with quoted evidence (Phase 5)
  8. Pre-trade skills vs. what happened (Phase 6)
  9. Process Score breakdown and the 2×2 matrix with the trade placed in its cell (Phase 8)
  10. The journal lesson as a paste-ready decision-log row (Phase 9)
  11. Re-entry view (Phase 10), Thesis Invalidation, and the Investment Signal block

Year mode — Data & Sources; a one-line-per-trade table (ticker · dates · return · excess · Process Score · quadrant · top bias); the aggregate statistics table (Phase 7) with the disposition-effect test spelled out; the quadrant counts in the 2×2; the year's one lesson; then the re-entry view, Thesis Invalidation and signal block for the named (or largest-loss) ticker.


Example

User: trade-postmortem ZEPH   (reads output/thesis/ZEPH.md, status CLOSED)

[ZEPH is a fictional ticker; all figures are hypothetical.]

The file's plan: thesis opened 2025-03-03 — gross margin expands past 60% as the new
product line scales. KPI gross margin ≥ 55%; exit trigger "GM < 55% for one quarter →
exit". Ladder: 100 sh at $40, 100 at $37, 100 at $34; ceiling 300 sh.

The decision log shows: bought 100 @ $40.00 and 100 @ $37.00 per plan; 2025-08-12 the
10-Q showed GM 53.9% and the status went WEAKENED, yet on 2025-08-14 the user bought
150 @ $33.00 ("great chance to lower my average"); the exit trigger FIRED 2025-11-06
(GM 52.1%) with the stock at $34; the user held ("I'll sell when it's back to $36") and
sold all 350 sh @ $29.00 on 2026-02-10. Fees $15, no dividends.

Outcome: cost $12,650, proceeds $10,150 → −$2,515 gross, −$2,530 after fees,
−20.0% on capital deployed. Over the window, SPY +9.0%, sector ETF +4.0% (hypothetical):
excess vs. sector −24.0 pp → bad outcome. Attribution (first entry $40 → exit $29,
−27.5%): market +9.0, sector −5.0, stock-specific −31.5 pp. Plan return: 100 @ $40 +
100 @ $37, no add while WEAKENED, exit 200 sh @ $34 when the trigger fired → −$910
(−11.8%). Timing effect: −$1,620 — the cost of the two deviations.

Plan adherence: entry 10, sizing 10, adds 0 (added while WEAKENED), ceiling 0 (350 > 300),
exit 0 (trigger lag ~65 trading days), reviews 5 (the August check ran 3 weeks late)
→ adherence 25 / 6 = 4.2. Biases: sunk cost confirmed (quote: "lower my average"), anchoring
confirmed ("back to $36" is near the $36.14 average cost), disposition suspected.
Pre-trade skills: stock-eval was wrong on the mechanism; bear-case thesis-killer #2
(competitor pricing) was right and was not wired into a trigger.

Process Score = 0.25×8 + 0.35×4.2 + 0.25×5 + 0.15×6 = 5.6 → bad-decision row (borderline)
→ Deserved loss. Lesson: "When the file says WEAKENED, the ladder is closed — I added
150 shares into a margin miss and it cost me $1,620 against my own plan."

Re-entry view at today's price: the margin mechanism was falsified, no fresh valuation
supplied → re-entry Score 3.5 / 10, BEARISH, Action SELL (stay out; no re-entry without
a new thesis). Then Thesis Invalidation and the signal block.

Notes

  • Grade the decision, not the price chart. The hindsight firewall (Phase 1) is the whole point of the skill; a post-mortem that uses later facts to grade earlier decisions teaches the outcome, which is exactly the lesson the 2×2 exists to prevent.
  • A good outcome never raises the Process Score, and a bad one never lowers it. If the user pushes back ("but it made money"), the answer is the dumb luck cell.
  • Benchmarks are total-return and window-matched. Price-only SPY understates the market by its dividend yield; a calendar-year benchmark against a seven-month trade is meaningless.
  • One lesson. Five lessons are a list nobody re-reads. Pick the one with the largest dollar cost or the most frequent bias.
  • Tax figures are pointers, not calculations. Short- vs. long-term status and wash-sale exposure on a re-entry go to tax-lens; this skill is educational only.
  • Before: thesis-tracker <TICKER> --close (so the file is CLOSED with its final log row). After: paste the lesson into the next thesis file's decision log; run learning-coach --quiz on lesson 13 if a bias was confirmed; and, if re-entering, thesis-tracker Open with a new thesis, position-ladder for the size, and tax-lens for the wash-sale window.
  • Keep closed files; year mode reads them. Where the environment cannot read output/thesis/, ask the user to paste the file.

Thesis Invalidation

After delivering the analysis signal, specify what would reverse it:

If signal is BULLISH (re-entry justified) — the view breaks if:

  • The KPI that the original thesis depended on (named in the old file) prints below its threshold again at the next filing — the post-mortem judged the exit a process error; a second breach says it was the thesis
  • The price rises above the top of the current stock-valuation range before a new thesis file is written — the re-entry becomes a chase, which is the FOMO pattern from Phase 5
  • The bias the post-mortem confirmed shows up in the new trade's first decision-log row (e.g. a size above the cap "because I know this one now")

If signal is BEARISH (stay out) — the view breaks if:

  • The mechanism the post-mortem judged falsified is restored by new evidence (e.g. two consecutive quarters back above the old KPI threshold, from the filings, not from guidance)
  • The price falls well below a freshly computed valuation range with the business stable — the "no" was about price, and the price has changed
  • The user writes a new, narrower thesis that does not depend on the falsified mechanism, and it survives bear-case

Re-run this analysis when:

  • Next earnings release of the ticker (for the re-entry view)
  • The user is about to re-enter the ticker — run it before the order, not after
  • Price moves ±15% from current level
  • Year end (year mode), or after every 10 closed trades
  • Material news event (acquisition, leadership change, regulatory decision)

Standard Signal Output

This skill grades a closed trade, so the signal block cannot describe it. State the mapping explicitly: the box expresses the re-entry view on the ticker today (Phase 10) — Signal: BULLISH means that, knowing what the post-mortem found, buying it again at today's price is justified under a new written thesis; NEUTRAL means no edge either way, or no fresh data (Score 5.0, Confidence LOW, Action HOLD); BEARISH means stay out. Action: BUY means "open a new thesis-tracker file and size with position-ladder" — never "re-open the old trade"; Action: SELL means "do not re-enter" (and close any residual shares). In year mode the box covers the ticker the user named, or the year's largest-loss trade.

All analysis concludes with this standardized block:

In the output, the box below is followed immediately by this machine-readable footer — the same values as the box, filled in (no placeholders, no comments), so tools can read the result without parsing the box:

{"investskill": "1", "skill": "trade-postmortem", "ticker": "<TICKER, or null for a portfolio / market-wide view>", "as_of": "<YYYY-MM-DD from Data & Sources>", "signal": "<BULLISH | NEUTRAL | BEARISH>", "confidence": "<HIGH | MEDIUM | LOW>", "horizon": "<SHORT-TERM | MEDIUM-TERM | LONG-TERM>", "score": 0.0, "action": "<BUY | HOLD | SELL>", "conviction": "<STRONG | MODERATE | WEAK>", "retrieval": "<pasted by user | web/tool retrieval | model memory>", "sources": ["<each primary source, dated>"]}
╔══════════════════════════════════════════════╗
║              INVESTMENT SIGNAL               ║
╠══════════════════════════════════════════════╣
║ Signal:      BULLISH / NEUTRAL / BEARISH     ║
║ Confidence:  HIGH / MEDIUM / LOW             ║
║ Horizon:     SHORT / MEDIUM / LONG-TERM      ║
║ Score:       X.X / 10                        ║
╠══════════════════════════════════════════════╣
║ Action:      BUY / HOLD / SELL               ║
║ Conviction:  STRONG / MODERATE / WEAK        ║
╚══════════════════════════════════════════════╝

Score Guide: 8.0–10.0 Strongly Bullish | 6.0–7.9 Moderately Bullish | 4.0–5.9 Neutral | 2.0–3.9 Moderately Bearish | 0.0–1.9 Strongly Bearish Confidence: HIGH (strong data, clear signals) | MEDIUM (mixed signals) | LOW (limited data, conflicting signals) Horizon: SHORT-TERM (1 week–3 months) | MEDIUM-TERM (3 months–1 year) | LONG-TERM (1+ years)

Note: The Score in the box is the Re-entry Score from Phase 10, mapped onto the standard scale. The Process Score is not the box Score — it grades past decisions and is reported in the body (Phase 8 and the verdict line); a high Process Score on a closed trade says nothing about whether to buy the stock again.

Disclaimer: Educational analysis only. Not financial advice.

Sponsors

Details

Marketplace
yennanliu/InvestSkill
Source
plugins/us-stock-analysis
Author
yennanliu
Category
Finance
Keywords
stock, investment, analysis, us-markets, finance, dcf, valuation, 10-k, 10-q, financial-reports, etf-analysis, tax-lens, earnings-preview, thesis-tracking, risk-stress-test, fact-check, forensic-accounting, credit-check, compare, trade-postmortem