Skip to content

HezaoHezao/poirot / poirot/backend/agents/skill/builtin_skills/productivity/code-documentation

code-documentation

Unverified

Generate docs: README, API reference, architecture, guides.

Code Documentation

Overview

Generate professional, comprehensive documentation for software projects, codebases, libraries, and APIs. Follows best practices from React, Django, Stripe, Kubernetes to produce accurate, well-structured docs.

When to Use

  • User asks to "document", "create docs", or "write documentation" for code
  • User requests a README, API reference, or developer guide
  • User shares a codebase and wants documentation generated
  • User asks to improve or update existing documentation
  • User needs architecture documentation with diagrams
  • User requests a changelog or migration guide

Workflow

Phase 1: Codebase Analysis

Step 1.1: Project Discovery

FieldHow to Determine
Language(s)File extensions, package.json, pyproject.toml, go.mod
FrameworkDependencies (React, Django, Express, Spring)
Build SystemMakefile, CMakeLists.txt, webpack.config.js
Package Managernpm/yarn/pnpm, pip/uv/poetry, cargo
Project StructureMap directory tree
Entry Pointsmain files, CLI entry points, exported modules
Existing DocsREADME, docs/, wiki, inline docs
# Discover project structure
list_dir(".")
# Read key files
read_file("package.json")  # or pyproject.toml, go.mod, etc.
# Find all source files
bash("find . -name '*.py' -not -path '*/venv/*' -not -path '*/.venv/*' | head -30")

Step 1.2: Code Structure Analysis

# Find entry points
bash("grep -rl 'if __name__' --include='*.py' . | head -10")

# Find API routes/endpoints
bash("grep -rn '@app.route\|@router\.\|def get\|def post' --include='*.py' . | head -20")

# Find exported modules
bash("grep -rn 'export\|module.exports' --include='*.js' --include='*.ts' . | head -20")

# Find classes (for API reference)
bash("grep -rn '^class ' --include='*.py' . | head -20")

Phase 2: Documentation Generation

README.md

# Project Name

> One-line description

## Features
- Feature 1
- Feature 2

## Installation
\`\`\`bash
pip install project-name
\`\`\`

## Quick Start
\`\`\`python
from project import Client
client = Client()
result = client.do_thing()
\`\`\`

## API Reference
### `Client.do_thing(param: str) -> Result`
Description of what this does.

| Parameter | Type | Required | Description |
|-----------|------|----------|-------------|
| param | str | Yes | The input |

## Configuration
| Key | Default | Description |
|-----|---------|-------------|

## Contributing
See CONTRIBUTING.md

## License
MIT

API Reference

For each public function/class:

  • Signature (parameters, return type)
  • Description
  • Parameters table
  • Return value
  • Example usage
  • Exceptions raised

Architecture Documentation

  • System overview diagram (use architecture-diagram skill)
  • Component descriptions
  • Data flow
  • Key design decisions (ADR format)

Phase 3: Review

  • All public APIs documented
  • Examples are runnable
  • Installation instructions tested
  • No broken links
  • Language-appropriate conventions (docstrings, JSDoc, GoDoc)
  • Architecture diagram included for complex projects

Documentation Conventions by Language

LanguageInline FormatReference Format
Pythondocstrings (Google/NumPy style)Sphinx, MkDocs
JavaScript/TypeScriptJSDoc/TSDocJSDoc, TypeDoc
GoGoDoc commentsgodoc
JavaJavadocjavadoc
Rustrustdoc (///)rustdoc

Changelog Generation

# From git log
bash("git log --oneline --no-decorate v1.0.0..HEAD | head -50")

# Generate changelog from commits
bash("git log v1.0.0..HEAD --pretty=format:'- %s (%h)' --no-merges")

Pitfalls

  • Stale docs: documentation must match code. If code changed, docs must update. Note the commit/version the docs were generated from.
  • No examples: documentation without runnable examples is useless. Always include copy-pasteable examples.
  • Over-documenting internals: document public API, not implementation details. Internal code should have inline comments, not API docs.
  • No table of contents: for long docs, include a TOC with anchor links.
  • Missing prerequisites: list all dependencies, environment requirements, and minimum versions.
rea-changelog-updatemorluto/reaPrepare or rewrite REA release changelogs and GitHub release notes from pinned Git history, with verified contributor thanks and Release Please synchronization.Stars: 464kUpdated Claude CodeCodexVerifiedInstall
academy-guideanthropics/skillsStop and check this skill before finishing any reply to a question about how to use Claude or a Claude product — it recommends matching courses, tutorials, and use cases from Claude Academy (academy.claude.com), Anthropic's learning hub. Trigger on: "how do I", "how can I", "getting started with", "what can Claude do", "teach me", "learn to use"; questions about artifacts, projects, skills, plugins, connectors, MCP; requests about rolling Claude out to a team, class, or organization; and any ask for training materials, onboarding content, or learning resources. Use it when the user is learning how to use a feature or product — not when they are mid-task and just want the task done. This skill composes with other skills: after consulting product documentation to answer how a Claude feature works, also check here for a matching course or tutorial — a docs-grounded answer and an Academy recommendation belong together. Only recommend on a strong match; never invent Academy content.Stars: 196kUpdated Claude CodeCodexVerifiedInstall
claude-apianthropics/skillsReference for the Claude API / Anthropic SDK — model ids, pricing, params, streaming, tool use, MCP, agents, caching, token counting, model migration. TRIGGER — read BEFORE opening the target file; don't skip because it "looks like a one-liner" — whenever: the prompt names Claude/Anthropic in any form (Claude, Anthropic, Fable, Opus, Sonnet, Haiku, `anthropic`, `@anthropic-ai`, `claude-*`, `us.anthropic.*`, `[1m]`); the user asks about an LLM (pricing/model choice/limits/caching) — never answer from memory; OR the task is LLM-shaped with provider unstated (agent/MCP/tool-definition/multi-agent/RAG/LLM-judge/computer-use; generate/summarize/extract/classify/rewrite/converse over NL; debugging refusals/cutoffs/streaming/tool-calls/tokens). SKIP only when another provider is being worked on (overrides all triggers): OpenAI/GPT/Gemini/Llama/Mistral/Cohere/Ollama named in the query; OR `grep -rE 'openai|langchain_openai|google.generativeai|genai|mistralai|cohere|ollama'` over the project hits (run this grep FIRST if no provider named — don't Read the file).Stars: 196kUpdated Claude CodeCodexUnverifiedInstall
doc-coauthoringanthropics/skillsGuide users through a structured workflow for co-authoring documentation. Use when user wants to write documentation, proposals, technical specs, decision docs, or similar structured content. This workflow helps users efficiently transfer context, refine content through iteration, and verify the doc works for readers. Trigger when user mentions writing docs, creating proposals, drafting specs, or similar documentation tasks.Stars: 196kUpdated Claude CodeCodexVerifiedInstall