New: 98,000+ skills indexed from GitHub repos with 100+ stars · every skill safety-checkedBrowse top skillsTop skills
Esc to closeBrowse all skills
Submit a skillBrowse skills
Submit a skillBrowse skills

Agent skill

Review Docs

by heypinchy in heypinchy/pinchy

Use before opening a PR that changes docs/ or a user-visible surface (an API route, the tool registry, an agent template, the audit event catalogue, the settings navigation, plugin tools), and when…

AGPL-3.0Auto-check passed

Install Review Docs

skills CLI
$ npx skills add heypinchy/pinchy --skill review-docs -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install heypinchy/pinchy review-docs --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/heypinchy/pinchy.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.claude/skills/review-docs .claude/skills/review-docs && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
review-docs
GitHub stars
182
Token cost
~1.3k tokens
SKILL.md length
689 words
Files
1
Skills in repo
18
Repo updated
First seen
Licence
AGPL-3.0

At a glance

Use before opening a PR that changes docs/ or a user-visible surface (an API route, the tool registry, an agent template, the audit event catalogue, the settings navigation, plugin tools), and when…

  • Works in 6 steps: Establish what actually changed → Find the prose that describes it → Read each hit against the code → …
  • Asks to review the docs
  • SKILL.md covers Why this exists, Scope, Steps and Before you finish, plus 1 more section
  • Calls pnpm, git and node

What it does

Review Docs is an agent skill from heypinchy/pinchy. Use before opening a PR that changes docs/ or a user-visible surface (an API route, the tool registry, an agent template, the audit event catalogue, the settings navigation, plugin tools), and when the user asks to "review the docs", "check the docs", or "are the docs still right". Reads the changed prose against the code it describes. Runs locally — the deterministic guards run in CI, this is the layer they cannot reach.

Its SKILL.md is about 1.3k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

The repository describes itself as: Self-hosted AI agent platform built on OpenClaw. Enterprise-ready, offline-capable, open source. 🦞. The licence is AGPL-3.0.

When your agent uses it

  • Asks to review the docs
  • Are the docs still right

Example prompts

  • “review the docs”
  • “check the docs”
  • “are the docs still right”
  • “/review-docs”

Workflow steps

6 steps, taken from the step headings in SKILL.md.

  1. Establish what actually changed
  2. Find the prose that describes it
  3. Read each hit against the code
  4. Check the direction nobody checks
  5. Voice
  6. Report

What it can do on your machine

Read from SKILL.md and the folder at commit 5159959. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Shell commands in SKILL.md call:

    • pnpm
    • git
    • node
    • gh

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md. Its commands use pnpm, git and gh, which can reach the network depending on how they are called.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Review Docs loads about 1.3k tokens when it runs. Until then it costs about 109 tokens; SKILL.md has 689 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~109
When it runs · the whole SKILL.md, loaded when a task matches
~1.3k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from heypinchy/pinchy at commit 5159959, republished under its AGPL-3.0 licence (© heypinchy). 689 words, ~1,338 tokens.

Download SKILL.mdSave it as .claude/skills/review-docs/SKILL.md (or your agent's skills folder).
name
review-docs
description
Use before opening a PR that changes docs/ or a user-visible surface (an API route, the tool registry, an agent template, the audit event catalogue, the settings navigation, plugin tools), and when the user asks to "review the docs", "check the docs", or "are the docs still right". Reads the changed prose against the code it describes. Runs locally — the deterministic guards run in CI, this is the layer they cannot reach.

Review the docs against the code

Why this exists

Three checks already run in CI and need no help from you:

GuardCatches
docs-coveragean API route / audit event / tool that no doc mentions
docs-consistencyan orphaned page, a Settings → X naming no tab, a promise with no issue
check-docs-requireda user-visible change with no docs/ change at all

Those find missing identifiers. They cannot read a sentence. The 2026-07-30 audit found four things that no identifier check could ever see:

  • A GET /api/settings/domain entry whose response fields were wrong — the path and method were both fine, the body was fiction.
  • A page that promised "a progress UI is planned for a later phase" while describing the shipped progress UI 120 lines earlier.
  • A table of "agent templates and default permissions" listing 2 of ~35 templates, with the wrong tools on the one row that mattered.
  • A live guide stating "Pinchy will not silently re-assign agents" about code that does exactly that.

Every one of those is a reading task. That is this skill.

Scope

Review the docs that describe what this branch changed — not the whole corpus. A full audit is a release-time activity (see cut-pinchy-release).

Steps

1. Establish what actually changed
bash
git diff --stat origin/main...HEAD

Separate the diff into behaviour (routes, tools, templates, permissions, audit events, UI labels, defaults, limits) and everything else. Only the first kind can falsify prose.

2. Find the prose that describes it

For each behaviour change, grep the docs for the thing itself and for its consequences — the second is where staleness hides:

bash
grep -rn "<identifier>" docs/src/content/docs/
grep -rni "<the user-facing noun>" docs/src/content/docs/

A rename is easy. What gets missed is the sentence three pages away that depends on the old behaviour without naming it.

3. Read each hit against the code

For every doc passage you found, open the code and check, in this order:

  1. Is the claim still true? Not "does it mention the right function" — does the described behaviour match what the code does.
  2. Are the specifics right? Response fields, status codes, defaults, limits, required-vs-optional, which side effects fire. These are wrong far more often than names are, and no guard checks them.
  3. Does the page contradict itself? Read the whole page, not the hunk. The contradiction is usually a "scope" or "limitations" section written earlier and never revisited.
  4. Does another page contradict this one? Concept pages and guides drift apart because they are edited by different changes.
Show full SKILL.md (291 more words)Show less
4. Check the direction nobody checks

The guards are one-directional: they ask "is everything in the code documented?" Ask the reverse: does the docs describe something the code no longer has? Grep the docs for the identifiers of anything this branch removed or renamed. Nothing in CI asks this question.

5. Voice

Read PERSONALITY.md before writing. English, "we", plain and specific. New prose should read like the page it lands in.

6. Report

For each finding: file and line, what the doc claims, what the code does, and the one-line fix. Rank by whether a reader would act wrongly on it.

If the changed behaviour is documented correctly and no page contradicts it, say so plainly — a clean review is a result, and inventing a finding to look thorough wastes the next person's time.

Before you finish

The deterministic gates are cheap; run them so the PR does not bounce:

bash
pnpm test:scripts && pnpm format:check
bash
cd docs && pnpm build && pnpm check:anchors && pnpm check:tables

Then record the review — this is what lets gh pr create through:

bash
node scripts/mark-docs-reviewed.mjs

The marker holds the current HEAD sha. Record it last, after any fix you made in response to the review: land another commit and the marker no longer matches, the hook fires again, and you review the new state. That is the intent, not an inconvenience.

If the honest answer is that the docs don't move, don't mark — put the reason in a commit trailer instead, where the next reader will find it:

Docs-not-needed: gateway-only ingress, no reader-facing path

The same trailer waives the CI gate, so one decision is recorded once.

What this skill is not

Not a rewrite. Do not restructure a page you were asked to check. If a page needs restructuring, say so and let the user decide — a docs PR that also reorganises is a docs PR nobody can review.

© heypinchy, AGPL-3.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in .claude/skills/review-docs of heypinchy/pinchy.

Open the folder on GitHubat commit 5159959

Compare with similar skills

Review Docs next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Review Docs compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Review Docs this skillheypinchy/pinchy182—~1.3kAutomated safety check: PassAGPL-3.0
Terminal Openeraffaan-m/ECC276k—~635Automated safety check: PassMIT
Open Federationruvnet/ruflo74k—~1.6kAutomated safety check: PassMIT
Open NotebookK-Dense-AI/scientific-agent-skills48k1 repos~2.8kAutomated safety check: PassMIT
Desk Opengithub/awesome-copilot40k—~757Automated safety check: PassMIT
Opendavepoon/buildwithclaude3.6k—~1.2kAutomated safety check: PassMIT

Similar skills

  • Terminal Opener

    affaan-m/ECC

    Open an executable and its argument array in a visible terminal window through a reusable, shell-free launch plan with dry-run, JSON, capability detection, detached fallback, and standalone recovery…

    276k GitHub stars~635 tokensUpdated 4 days ago
    Marketing & SEOAuto-check passed

  • Open Federation

    ruvnet/ruflo

    Coordinate with the open ruflo swarm federation at x.ruv.io (signed Nostr, membership-gated) and ask Seraphina — the swarm queen / primary coordinator — for guidance.

    74k GitHub stars~1.6k tokensUpdated today
    Auto-check passed

  • Open Notebook

    K-Dense-AI/scientific-agent-skills

    Organizes research with the self-hosted Open Notebook alternative to NotebookLM.

    48k GitHub starsUsed in 1 repo~2.8k tokens
    AI & LLM EngineeringAuto-check passed

  • Desk Open

    github/awesome-copilot

    Official

    Create and open a new desk in the workshop. An agent skill from github/awesome-copilot.

    40k GitHub stars~757 tokensUpdated today
    Productivity & AutomationAuto-check passed

  • Open

    davepoon/buildwithclaude

    Boot the session, load context from the vault, and surface what matters

    3.6k GitHub stars~1.2k tokensUpdated today
    Auto-check passed

  • Open Sea Automation

    ComposioHQ/awesome-claude-skills

    Automate Open Sea tasks via Rube MCP (Composio). An agent skill from ComposioHQ/awesome-claude-skills.

    77k GitHub starsUsed in 3 repos~734 tokens
    Productivity & AutomationAuto-check passed

More from heypinchy/pinchy

All 18 skills in this repo

  • Knowledge Search

    heypinchy/pinchy

    Answer questions from the organization's indexed documents using knowledgesearch, and cite every claim back to a retrieved passage.

    182 GitHub stars~1.4k tokensUpdated 18 days ago
    Auto-check passed

  • Odoo Read

    heypinchy/pinchy

    Query and summarize data from a connected Odoo instance with the odoo read tools (describe, count, read, aggregate).

    182 GitHub stars~965 tokensUpdated 18 days ago
    Auto-check passed

  • Update Dependencies

    heypinchy/pinchy

    A skill your agent uses when bumping general npm/pnpm dependencies across the Pinchy workspace (root, packages/web, packages/plugins/, docs), when the user asks to "update dependencies," "check for…

    182 GitHub stars~1.5k tokensUpdated 18 days ago
    Auto-check passed

  • Update Ollama Cloud Models

    heypinchy/pinchy

    A skill your agent uses when a new Ollama Cloud model is announced or available (e.g.

    182 GitHub stars~3.9k tokensUpdated 18 days ago
    Auto-check: notes

  • Update Openclaw

    heypinchy/pinchy

    A skill your agent uses when bumping the pinned OpenClaw core version (openclaw npm package), when preparing a Pinchy release, or when the user asks to "update OpenClaw" / "upgrade OpenClaw" / check…

    182 GitHub stars~2.8k tokensUpdated 18 days ago
    Auto-check passed

  • Cut Pinchy Release

    heypinchy/pinchy

    A skill your agent uses when cutting, tagging, or publishing a new Pinchy version — e.g.

    182 GitHub stars~11k tokensUpdated 18 days ago
    Auto-check: notes

Questions about Review Docs

What does Review Docs do?

Use before opening a PR that changes docs/ or a user-visible surface (an API route, the tool registry, an agent template, the audit event catalogue, the settings navigation, plugin tools), and when…. Review Docs is an agent skill from heypinchy/pinchy. Use before opening a PR that changes docs/ or a user-visible surface (an API route, the tool registry, an agent template, the audit event catalogue, the settings navigation, plugin tools), and when the user asks to "review the docs", "check the docs", or "are the docs still right".

When should I use Review Docs?

Review Docs fits situations like: asks to review the docs; are the docs still right.

How do I install Review Docs in Claude Code?

Run `npx skills add heypinchy/pinchy --skill review-docs -a claude-code`. Or copy the skill folder (.claude/skills/review-docs in heypinchy/pinchy) into .claude/skills/review-docs in your project. Claude Code loads it when a task matches its description.

How do I install Review Docs in Codex?

Run `npx skills add heypinchy/pinchy --skill review-docs -a codex`. Or copy the skill folder (.claude/skills/review-docs in heypinchy/pinchy) into .agents/skills/review-docs in your project. Codex loads it when a task matches its description.

Can I use Review Docs in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add heypinchy/pinchy --skill review-docs -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/review-docs, .gemini/skills/review-docs, .github/skills/review-docs and .opencode/skills/review-docs in your project.

What does Review Docs need to run?

Going by SKILL.md and its folder, Review Docs needs the command-line tools its instructions call (pnpm, git, node and gh).

Does Review Docs access the network?

SKILL.md contains no URLs. Its commands use git and gh, which can reach the network depending on how they are called. This is read from the text; nothing was executed.

Is Review Docs safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Review Docs use?

Review Docs is published under the AGPL-3.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Review Docs use?

About 1.3k tokens (SKILL.md is roughly 5.4k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Review Docs?

Skills that share tags, products or a category with Review Docs: Terminal Opener (affaan-m/ECC, 276k stars), Open Federation (ruvnet/ruflo, 74k stars), Open Notebook (K-Dense-AI/scientific-agent-skills, 48k stars) and Desk Open (github/awesome-copilot, 40k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Review Docs?

heypinchy (a GitHub organization) maintains it in heypinchy/pinchy, which has 182 GitHub stars. The repository holds 18 skills in this directory. The repository was last updated on September 21, 2026.

Source: heypinchy/pinchy on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.