Claude Marketplaces
SkillsMCP serversMarketplacesDigestTools
Submit

This week in Claude

Every Monday: Claude Code, Agent SDK, MCP, and the Anthropic platform moves worth your time.

Skills by category
Frontend DevelopmentBackend & APIsTesting & QASecurityDevOps & CI/CDGit & Pull RequestsDocumentationCode Review & QualityAI & Agent BuildingSkill Development
MCP servers by category
Sales & MarketingWeb & Browser AutomationDatabasesAI & LLM ToolsCloud & InfrastructureCommunication & MessagingDeveloper ToolsDesign & CreativeDocuments & KnowledgeSearch & Web Crawling
Marketplaces by category
AI Agents & OrchestrationLLM IntegrationDevelopment ToolsFrontend & UIBackend & APIsDatabasesTesting & Code QualityDevOps & CloudSecurity & ComplianceGit & Version Control
Claude Marketplaces

An independent directory of skills, plugins and MCP servers for Claude Code. Not affiliated with Anthropic.

Browse

  • Skills
  • MCP servers
  • Marketplaces
  • Weekly digest
  • Skill index
  • MCP index
  • Marketplace index

Site

  • About
  • Tools
  • Submit
  • Feedback
  • Privacy
  • llms.txt
  • Plugins reference

Maker

  • Snug
  • Agent Caffeine
  • knobs
  • Wax
  • Vectoys
  • mertduzgun.com
Crafted at
Skills/Documentation
hoangnguyen0403 avatar

Dev Fix

hoangnguyen0403/agent-skills-standard

This is a structured bug-fix pipeline that walks you through the full cycle: pull the ticket context from Jira or GitHub, cross-reference docs and code, write an implementation plan artifact that requires your approval before any code gets written, then run the actual fix in a TDD loop, verify it locally with Playwright or Appium (not just "it compiles"), and finally generate the commit and PR with a walkthrough artifact showing your test evidence. It's opinionated about stopping for approval and proving the fix works before you push. Use it when you want guardrails around the fix process or need to document verification steps for compliance or team review. The artifact templates and hard stops make it slower but more auditable than just hacking a fix directly.

GitHub stars
504
last updated
Jun 11, 2026
View on GitHubSave

Install

$ npx -y skills add hoangnguyen0403/agent-skills-standard --skill dev-fix --agent claude-code

Installs into .claude/skills of the current project.

Files

Dev Fix Skill

[!IMPORTANT] Unified developer workflow for fixing bugs. Analyzes issue-tracker context, cross-checks docs/code, proposes a solution, implements the fix, verifies locally, and delivers a PR/MR.

Optional args: slug=, ticket=<id/url>, mode=interactive|autonomous|channel, channel=, auto_continue=true|false, profile=business|hybrid|technical.

Instructions

When the user asks to perform this workflow, execute the following steps:

Dev-Fix — Bug Remediation

Goal: Take a bug ticket from root-cause analysis through a locally-verified PR/MR, enforcing a strict Propose -> Approve -> Verify cycle.

Input

/dev-fix <issue-url-or-key>

Workflow

Step 0: Environment Prep (Turbo)

// turbo

  1. Sync Registry: git pull origin main in the standard repo.

Step 1: Research & Discovery (Implementation Plan Phase)

[!TIP] Sub-Agent Delegation: If your platform supports sub-agents, delegate ticket extraction to specialist-jira-analyst and context lookup to specialist-codebase-scout. If sub-agents are NOT supported, execute these steps yourself.

  1. Analyze Ticket: Use installed Jira/GitHub/GitLab/ADO MCP first; otherwise use exported ticket text. Extract Reproduce steps, Expected Result, and Actual Result.
  2. Cross-Check Context: Use knowledge-base MCP when configured; otherwise use local code search. Locate relevant code.
  3. Create Implementation Plan:
    • Use the Implementation Plan Template below.
    • Initialize project-local docs/prd/prd-plan-[slug].md.
    • Goal: Clear description of the root cause.
    • Proposed Changes: Exact files and logic to be modified.
    • Verification Plan: Detail which QE skill will be used to verify the fix locally before PR.
    • Do not propose code changes until repro steps, expected result, and root cause hypothesis are explicit.
    • SNC: score per common-task-complexity-routing; record snc_tier/model_tier in the plan.
  4. HARD STOP: Request user approval for the implementation plan. Mandatory at snc_tier=high; medium requires self-review first; low may skip approval in autonomous mode only.
  5. Readiness Gate: Run implementation-readiness; code only after READY or approved PARTIAL.

Step 2: Implementation (TDD Phase)

[!TIP] Sub-Agent Delegation: For the actual fix, delegate the TDD loop to specialist-tdd-implementer. If sub-agents are NOT supported, execute the TDD loop yourself using the TDD skill matched in AGENTS.md.

  1. Worktree Branching: Create a new worktree for the fix using git worktree add ../<ticket-key> -b fix/<ticket-key> and cd into it.
  2. Task Tracking:
    • Use the Task Template below.
    • Initialize project-local docs/srs/srs-task-list-[slug].md.
  3. Code: Implement the fix using common-tdd or the @specialist-tdd-implementer sub-agent. Follow common-best-practices and service-specific AGENTS.md rules.
    • Use common-tdd: strict RED first for new behavior; for legacy fixes, characterize only when needed and reproduce the intended change as RED without deleting unrelated implementation.
    • Record the Test Intent Record and run the smallest foreground, single-run target with a project timeout or 120-second fallback before escalating.

Step 3: Local Verification (Enterprise Standard)

Do NOT rely on "it builds" — verify the fix against the issue reproduction steps.

  1. Launch Dev Server: Run the local dev environment for the service.
  2. Execute QE Audit:
    • Web: Load quality-engineering-playwright-cli. Run its preflight, take the first driver rung (CLI, else Playwright MCP). Run the reproduction steps. Capture "After" snapshots.
    • Mobile: Load quality-engineering-appium-mcp. Run its preflight, take the first driver rung (local device, else cloud). Run the reproduction steps.
  3. Final Verdict: Compare results against the issue Expected Result. If any sub-3px regressions exist, fix them now.
    • No success claim without fresh local evidence in docs/srs/srs-walkthrough-[slug].md.

Step 4: Deliver PR

  1. Commit: Generate a commit message using caveman-commit.
  2. PR/MR Details: Draft provider-appropriate PR/MR notes and link the source issue.
  3. Walkthrough:
    • Use the Walkthrough Template below.
    • Create project-local docs/srs/srs-walkthrough-[slug].md with evidence of the local verification.

Runtime Contract

  • Use for bug tickets that need root-cause remediation and a PR/MR.
  • Required inputs: issue URL/key or exported ticket text with reproduce steps.
  • Accepts a test-loop REAL_BUG_DO_NOT_HEAL handoff as the ticket input: the healer's TEST: and EVIDENCE: lines plus the failing AC are the reproduce steps.
  • Return BLOCKED only when repro steps, expected result, or root-cause hypothesis cannot be established.

Handoff Payload

  • slug, operator_profile (carried, not re-inferred), snc_tier, model_tier, implementation plan path, task list path, walkthrough path, PR/MR link, outcome report, next workflow.

Blocking Questions

  • Ask max 3 at a time with a recommended default and 2-3 options.

Artifact Templates

Implementation Plan Template

# Implementation Plan: [Name]

## Goal

## Proposed Changes

## Task Slices

| Slice   | Scope   | Verification   |
| ------- | ------- | -------------- |
| [slice] | [scope] | [verification] |

## Risks

## Verification Plan

## Next Workflow
implementation-readiness

Task Template

# Task: [Name]

## Scope

## Checklist

- [ ] [task]

## Decisions

## Evidence

## Next Workflow
verify-work

Walkthrough Template

# Walkthrough: [Name]

## Scope

## Acceptance Criteria

## Evidence

| Check   | Result              | Evidence   |
| ------- | ------------------- | ---------- |
| [check] | [PASS/FAIL/BLOCKED] | [evidence] |

## Risks

## Outcome Report
{schema_version: 1, run_id: "[run-id]", slug: "[slug]", workflow: dev-fix, feature_status: implemented, started_at: "[timestamp]", completed_at: "[timestamp]", requirement_trace: {brd_objectives: [], requirements: [], acceptance_criteria: [], srs: []}, completed_evidence: [], missing_evidence: [], decision_needed: [], recommended_next_workflow: verify-work, cost: {source: unavailable}, agent: {identity: "[agent-identity]", model: "[model]"}}

## Next Workflow
verify-bug | verify-work

Cost Report

Call get_session_cost(workflow="dev-fix") before final handoff.

Anti-Patterns

  • No Blind Implementation: Never write code before the implementation plan is approved.
  • No Orphan Sessions: Always close browser/appium sessions used during verification.
  • No skipping local verify: "I checked it manually" is not enough. Provide snapshots/logs in the project-local docs/srs/srs-walkthrough-[slug].md.
SKILL.md on GitHub

Sponsors

SnugThe clipboard manager you and your agent can use, safelyAgent CaffeineClose your lid, come back to finished workWaxBlock anything with supervision, without erasing your iPhoneOpen StepsSponsoredFree and open-source skills that make AI coding agents easier to understand, verify, and control.

Details

Repository
github.com/hoangnguyen0403/agent-skills-standard
Categories
Documentation, AI & Agent Building, Debugging, Automation & Workflows
License
Apache-2.0
First seen
Jun 11, 2026

More from hoangnguyen0403/agent-skills-standard

All 28
  • Implement Feature
  • Implementation Readiness
  • Pentest
  • Plan Feature
  • Publish Notes
  • Retro Learn
  • Review Ticket
  • Sdlc
  • Security Test
  • Session Report
  • Skill Benchmark
  • Skill Creator
  • Traceability Audit
  • Update Docs
  • Verify Bug
  • Verify Work
  • Zephyr Coverage Analysis
  • Battle Test
  • Brainstorm Feature
  • Caveman
  • Caveman Commit
  • Caveman Compress
  • Caveman Review
  • Code Review

Recommended

More Documentation
1hoangsonww avatarrelease-guardhoangsonww/claude-code-agent-monitorRun release-readiness checks for this repository. Use when validating docs, scripts, verification coverage, and operational safety before merge or release.0installs
2hoangsonww avatarship-featurehoangsonww/claude-code-agent-monitorImplement a feature safely end-to-end in this repository. Use when adding or changing functionality across backend, frontend, or MCP with required verification 0installs
3hoodini avatarvideo-edithoodini/ai-agents-skillsEdit any video into a captioned showcase — transcribe (any language, defaults to large-v3), present a transcript_review.txt for the user to fix mishears BEFORE 0installs
4htmlstreamofficial avatarpreline-theme-generatorhtmlstreamofficial/prelineGenerate, preview, save, or validate Preline theme CSS using the bundled local generator in this skill. Use when the user asks to create a new Preline theme, ma0installs
5https-deeplearning-ai avatarlearning-a-toolhttps-deeplearning-ai/sc-agent-skills-filesCreate learning paths for programming tools, and define what information should be researched to create learning guides. Use when user asks to learn, understand0installs
6huangjia2019 avatarapi-conventionshuangjia2019/claude-code-engineeringAPI design patterns and conventions for this project. Covers RESTful URL naming, response format standards, error handling, and authentication requirements. Use0installs