review-docs
Review and clean technical documentation (Markdown/README/runbooks/ADRs). Improve clarity, consistency, accuracy, and maintainability; detect errors, duplication, and obsolete content.
Best use case
review-docs is best used when you need a repeatable AI agent workflow instead of a one-off prompt.
Review and clean technical documentation (Markdown/README/runbooks/ADRs). Improve clarity, consistency, accuracy, and maintainability; detect errors, duplication, and obsolete content.
Teams using review-docs should expect a more consistent output, faster repeated execution, less prompt rewriting.
When to use this skill
- You want a reusable workflow that can be run more than once with consistent structure.
When not to use this skill
- You only need a quick one-off answer and do not need a reusable workflow.
- You cannot install or maintain the underlying files, dependencies, or repository context.
Installation
Claude Code / Cursor / Codex
Manual Installation
- Download SKILL.md from GitHub
- Place it in
.claude/skills/review-docs/SKILL.mdinside your project - Restart your AI agent — it will auto-discover the skill
How review-docs Compares
| Feature / Agent | review-docs | Standard Approach |
|---|---|---|
| Platform Support | Not specified | Limited / Varies |
| Context Awareness | High | Baseline |
| Installation Complexity | Unknown | N/A |
Frequently Asked Questions
What does this skill do?
Review and clean technical documentation (Markdown/README/runbooks/ADRs). Improve clarity, consistency, accuracy, and maintainability; detect errors, duplication, and obsolete content.
Where can I find the source code?
You can find the source code on GitHub using the link provided at the top of the page.
Related Guides
SKILL.md Source
Act as a Technical Writer + Senior Engineer + QA. Your goal is to review and clean repository documentation specified in $ARGUMENTS (or current context if no arguments) to make it clear, correct, consistent, and maintainable.
Deliver in this format:
A) SUMMARY
- Status: ✅ Ready / ⚠️ Requires adjustments / ❌ Inconsistent or dangerous
- Top 5 issues (prioritized)
- Minimum actions to reach "✅ Ready"
B) FINDINGS (prioritized)
For each finding include:
- Severity: P0 (blocking) / P1 / P2 / P3
- Evidence: file:section (or exact heading)
- Problem: what's confusing or wrong
- Proposed fix: suggested text or concrete restructuring (in Markdown)
C) REWRITE PROPOSAL (if applicable)
- Proposed index (TOC) or recommended structure
- Sections to merge/delete/move
- List of normalized "names/terminology"
Review and cleanup criteria:
1) Accuracy and currency
- Detect obsolete content (commands, paths, flags, dependencies, versions, processes).
- Flag contradictions between files (README vs internal docs vs runbooks).
- Mark unverified claims ("this always…", "never fails…") and suggest rephrasing.
2) Clarity and readability
- Long sentences, ambiguities, logical jumps.
- Rewrite so a new developer understands the "what", "why", and "how".
- Add minimum context: prerequisites, limits, gotchas.
3) Editorial and technical consistency
- Unify terminology, component names, capitalization, list style, verb tenses.
- Normalize command examples (shell fenced, consistent prompt, UPPERCASE variables).
- Maintain convention: "imperative" for steps ("Execute…", "Verify…").
4) Security and compliance
- Find and remove/anonymize secrets, tokens, credentials, sensitive internal URLs, or PII.
- Avoid recommending insecure practices (e.g., "disable TLS", "chmod 777", "export AWS_SECRET…").
- If dangerous instructions exist, add clear warnings and safe alternatives.
5) Operations and runbooks
- Verify runbooks have: symptoms → diagnosis → mitigation → rollback → verification.
- Add "before/after" checks and measurable success criteria.
- Flag non-deterministic steps or those dependent on tribal knowledge.
6) Actionability
- Each section must allow executing the task without guessing:
- prerequisites
- concrete commands
- examples of expected inputs/outputs
- relevant internal links (no broken links)
Rules:
- Don't invent tools/processes: if data is missing, mark "NEEDS CONFIRMATION" and propose what to ask or where to verify in the repo.
- Minimize meaning changes: prioritize clarity and correctness, not style for style's sake.
- When proposing text, deliver it ready to paste into Markdown.Related Skills
review-leaks
Detect secrets, credentials, and sensitive data leaks before pushing to public repositories.
review-for-prod
Production-ready Go code review (QA + security + maintainability) for this project only.
release
Create a professional release using GitHub CLI (gh). Generate SemVer version, clear release notes, and ready-to-run command.
flutter-dart-code-review
库无关的Flutter/Dart代码审查清单,涵盖Widget最佳实践、状态管理模式(BLoC、Riverpod、Provider、GetX、MobX、Signals)、Dart惯用法、性能、可访问性、安全性和整洁架构。
security-review
Use this skill when adding authentication, handling user input, working with secrets, creating API endpoints, or implementing payment/sensitive features. Provides comprehensive security checklist and patterns.
microsoft-docs
Query official Microsoft documentation to find concepts, tutorials, and code examples across Azure, .NET, Agent Framework, Aspire, VS Code, GitHub, and more. Uses Microsoft Learn MCP as the default, with Context7 and Aspire MCP for content that lives outside learn.microsoft.com.
addressing-pr-review-comments
Address all valid review comments on a PR for the current branch in the streamlit/streamlit repo. Covers both inline review comments and general PR (issue) comments. Use when a PR has reviewer feedback to address, including code changes, style fixes, and documentation updates.
lightning-architecture-review
Review Bitcoin Lightning Network protocol designs, compare channel factory approaches, and analyze Layer 2 scaling tradeoffs. Covers trust models, on-chain footprint, consensus requirements, HTLC/PTLC compatibility, liveness, and watchtower support.
gha-security-review
Find exploitable vulnerabilities in GitHub Actions workflows. Every finding MUST include a concrete exploitation scenario — if you can't build the attack, don't report it.
gh-review-requests
Fetch unread GitHub notifications for open PRs where review is requested from a specified team or opened by a team member. Use when asked to "find PRs I need to review", "show my review requests", "what needs my review", "fetch GitHub review requests", or "check team review queue".
fix-review
Verify fix commits address audit findings without new bugs
error-debugging-multi-agent-review
Use when working with error debugging multi agent review