petekp/agent-skills

Research a UI design aesthetic (brutalist, glassmorphism, cyberpunk, Apple HIG, neumorphism, Swiss modernist, Material Design, art deco, vaporwave, etc.) and produce exhaustive, implementation-ready design guidelines with exact CSS values, color palettes, component states, animations, and typography

Compile agent-optimized changelog by cross-referencing git history with plans and documentation to identify key decisions and stale information

Make application behavior visible to coding agents by exposing structured logs and telemetry endpoints. Guides adding structured logging, exposing logs for agent access, and documenting how to query them.

Write clear, plain-spoken code comments and documentation optimized for both human readers and AI coding assistants

Execute architectural refactoring from an assessment document with deterministic, chunked operations and aggressive verification at every step. Designed to prevent agent drift across sessions and context compactions.

Build a compilable type-level skeleton from a high-level architecture spec before writing any implementation logic. Proves architecture soundness via compiler verification before migration begins.

Assess a codebase's readiness for autonomous agent development and provide tailored recommendations

Interview-driven blog post drafting for technical product audiences. Use when user wants to write a blog post, article, or essay and needs help developing their thesis, structure, and initial draft. Triggers on "write a blog post", "draft an article", "help me write about X", "blog drafter", or when user has a topic they want to turn into written content. Conducts structured interviews using AskUserQuestion to extract the user's unique insights before generating drafts.

Capture learnings from sessions to project knowledge files or skills for cross-project reuse

Meta-cognitive decision support that analyzes context and surfaces intelligent next-step options via AskUserQuestion with recommendations

Generate a pedagogically-grounded study guide for learning an unfamiliar codebase using evidence-based learning techniques

Identify non-obvious signals, hidden patterns, and clever correlations in datasets using investigative data analysis techniques. Pairs with personality-profiler for enhanced signal extraction.

Remove LLM-isms and AI writing patterns from text to restore natural human voice

Systematically identify and remove dead code, redundant implementations, and orphaned artifacts in agent-maintained codebases

Conduct exhaustive, citation-rich research on any topic using all available tools: web search, browser automation, documentation APIs, and codebase exploration. Prioritizes accuracy over speed, cross-references claims across sources, identifies conflicts, and provides full citations with confidence levels.

Write changelog entries for open source documentation sites using Keep a Changelog format by reviewing git commits since the last entry

Systematic audit methodology for rooting out bugs, race conditions, stale docs, and correctness issues. Use for code audits, reviewing agent-written code, or deep analysis of any system.

Create high-quality animated explainer visuals for essays and blog posts. Use when visualizing concepts, processes, data, or ideas with interactive web animations. Outputs self-contained HTML/CSS/JS.

Build interactive debugging interfaces that reveal internal system behavior through visualization. Use for state machines, data flow, events, algorithms, render cycles, animations, or any mechanism with hidden internals.

Cold, methodical diagnostician for when you're stuck in agent-assisted app development. Diagnoses agent loops, architectural rot, environment corruption, and requirements drift, then intervenes precisely to unblock you.

Continuous formal verification of architectural constraints, behavioral invariants, and code elegance for agent-driven codebases

Generate a smart bootstrap prompt to continue the current conversation in a fresh session from where you left off

Expertise in architecting, implementing, reviewing, and debugging hierarchical matching systems including two-sided matching (Gale-Shapley), assignment/optimization (Hungarian algorithm), multi-level hierarchy matching, and entity resolution across hierarchies.

Apply interaction design principles for intuitive, responsive interfaces including micro-interactions, loading states, transitions, and accessibility patterns

Transform a codebase study guide into a polished interactive web experience with scroll-driven storytelling, interactive architecture diagrams, and animated code walkthroughs

Create a narrative guide to a codebase or feature in the style of Knuth's Literate Programming — code and prose interwoven as an essay, ordered for human understanding

Design and build native macOS applications with SwiftUI or AppKit following Apple's design guidelines and Liquid Glass design system

Guide users step-by-step through manually testing current work with selectable outcome options

Apply Model-First Reasoning (MFR) to code generation tasks with formal modeling before implementation

Synthesize outputs from multiple AI models into a verified assessment by cross-referencing claims against actual source code to resolve contradictions and produce a more reliable document than any single model

Evaluate and optimize agent documentation in dotfiles (.claude, .codex, .cursor, .aider) for better discoverability and retrieval efficiency

Product analytics expert using PostHog MCP. Surface insights, create dashboards, analyze funnels, track metrics, set up experiments, and understand user behavior with persistent product context.

CAVEMAN HUNT BAD PROCESS! Me find greedy creature eating fire (CPU) and rocks (memory). Me bonk them. Lightning rock (battery) happy. Tribe proud.

Facilitate methodical review of proposals by chunking content, predicting reviewer reactions, and producing feedback adapted to the proposal's format

Enter todo recording mode to capture ideas without acting on them, then organize and prioritize on exit

Craft a high-quality prompt for a deep research agent (like ChatGPT Deep Research) through adaptive interviewing. Turns vague research needs into precise, comprehensive prompts that get dramatically better results.

Create a comprehensive review package of your current work to hand off to another LLM for code review, architecture review, or both

Robust Rust patterns for file-backed data, parsing, persistence, FFI boundaries, and system integration

Ruthlessly analyze architectural seams to expose coupling problems, abstraction leaks, and design failures. Produces uncompromising redesign proposals that prioritize correctness over backwards compatibility.

Guided Shape Up workflow for taking projects from idea to working software. Orchestrates shaping and breadboarding through Frame, Shape, Breadboard, Slice, Build phases.

First-principles simplification analysis for codebases. Inventories what code actually does, then assesses whether the implementation is the simplest way to achieve it.

Create bespoke parameter tuning panels for iterating on animations, layouts, colors, typography, physics, or any visual parameters

Apply professional typography principles for readable, hierarchical, and aesthetically refined interfaces

Expert Unix and macOS systems engineer for shell scripting, system administration, command-line tools, launchd, Homebrew, networking, and low-level system tasks