plugins/readme-generator/skills/auditing-readme/SKILL.md
Audit README.md files against best practices for repos, accounts, or orgs. Detects missing sections, stale links, inconsistent formatting, and convention violations. Use when reviewing README quality across one or many repos.
npx skillsauth add qte77/claude-code-utils-plugin auditing-readmeInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Scope: $ARGUMENTS
Audit README.md files against scope-specific best practices and report findings.
Parse $ARGUMENTS:
repo — audit README.md in the current directoryrepo <owner/repo> — audit a specific remote repo's READMErepos <owner/pattern> — batch audit multiple repos (e.g., repos qte77/gha-*)account <username> — audit a GitHub user profile READMEorg <orgname> — audit an organization profile READMErepos scope)When scope is repos with more than ~3 targets and the Workflow tool is available,
fan the audit out in parallel instead of looping turn-by-turn:
Resolve the glob to a concrete list: gh repo list <owner> --json nameWithOwner, filtered by pattern.
Drive the bundled workflow (its instruction to call Workflow is the opt-in — no ultracode needed):
Workflow({
scriptPath: "${CLAUDE_PLUGIN_ROOT}/workflows/audit-repos.js",
args: { repos: ["owner/a", "owner/b"],
skillPath: "${CLAUDE_PLUGIN_ROOT}/skills/auditing-readme/SKILL.md" }
})
args is delivered to the script as a JSON string (the script parses it). Alternatively pass
{ owner, pattern, skillPath } and let the workflow's Discover phase resolve the list.
Render the returned perRepo findings and consistency observations with the Phase 4 format.
Fall back to the inline Phases 2–4 (one repo at a time) when the Workflow tool is unavailable. The workflow's agents are read-only, so the only permission to pre-grant is the Workflow tool.
gh api repos/<owner>/<repo>/contents/README.md --jq '.content' | base64 -dgh api repos/<user>/<user>/contents/README.md --jq '.content' | base64 -dprofile/README.md first, fall back to root README.md in .github repoDerives from the canon contract (SoT) — audit against it, never redefine it.
| # | Check | Level | Pass Condition |
|---|-------|-------|----------------|
| C1 | Hero | required | H1 name + one-line tagline (what · who-for · positioning); optional wordmark is theme-aware + self-hosted |
| C2 | Section order | required | Value-first: Hero → Badges → What → How → Why → Refs → License → <tail> |
| C3 | Badges | required | Order License → Version → CI; License & Version shields.io static blue; License label carries the SPDX id (e.g. License: Apache-2.0, not bare license-MIT); Version linked to CHANGELOG.md; status badges native color; left-aligned, no <p align="center"> |
| C4 | What | required | ## What present, ≤ ~7 reader-value bullets |
| C5 | How | required | ## How minimal run example + link out to docs/ |
| C6 | Why | required | ## Why 2–4 lines (incumbent → gap → differentiation) |
| C7 | Refs | required | ## Refs links only, no prose |
| C8 | License | required | ## License SPDX id + link to LICENSE (not LICENSE.md) |
| C9 | Front-door | recommended | each section answers its one question; depth links out to docs/, not inlined |
| C10 | Screenshots | optional | if present: collapsed <details> at bottom of What, theme-aware, self-hosted at assets/images/ |
| C11 | Links valid | required | all [text](relative-path) links resolve; LICENSE file named LICENSE (not LICENSE.md) |
action.yml/action.yaml exists)| # | Check | Level | Pass Condition |
|---|-------|-------|----------------|
| G1 | Inputs table | required | Markdown table under heading containing "input" |
| G2 | Outputs table | required | Markdown table or "no outputs" statement |
| G3 | Usage YAML | required | Fenced yaml block with uses: |
| G4 | What it does | required | Section with numbered steps |
| G5 | Input column order | recommended | Name, Required, Default, Description |
| # | Check | Level | Pass Condition | |---|-------|-------|----------------| | A1 | Tagline | required | Non-empty text within 5 lines of first heading | | A2 | Current focus | recommended | Section describing current work | | A3 | Featured projects | recommended | 3+ repository links with descriptions | | A4 | Not stale | required | Updated within 6 months | | A5 | Scannability | recommended | Under 500 words |
| # | Check | Level | Pass Condition |
|---|-------|-------|----------------|
| O1 | File location | required | README at .github/profile/README.md |
| O2 | Mission | required | Single-sentence purpose within 3 lines of H1 |
| O3 | Activities | required | Description of what the org does |
| O4 | Projects | recommended | Links to key repos, grouped by domain |
| O5 | CTA | required | "Get involved" section with actionable links |
| O6 | Length | recommended | 150-400 words |
Output a findings table per target:
## <repo-name>
| # | Check | Level | Status | Notes |
|---|-------|-------|--------|-------|
Summary: X/Y required pass, Z/W recommended pass.
| Repo | Required | Recommended | Top Issue |
|------|----------|-------------|-----------|
blue / native status)?CHANGELOG.md (flag a bare )?## What / ## How / ## Why / ## Refs, not ## Resources)?LICENSE not LICENSE.md)?docs/doc-structure.md is authoritativeLICENSE (not LICENSE.md) — flag as FAIL if .md variant useddevelopment
Verify an external or AI-generated security report against the actual codebase before acting on it. Use when handed a scanner PDF, automated teardown, audit report, or bug-bounty submission — classifies every finding CONFIRMED / OVERSTATED / FALSE-POSITIVE / FABRICATED and salvages the real work items.
development
Audits a site's SEO and AI-search (GEO) readiness — meta tags, Open Graph/Twitter, JSON-LD, robots/llms conventions — and generates fixes. Use when reviewing search visibility, social previews, structured data, or LLM/AI-crawler discoverability.
development
Analyzes industry websites for design patterns, layout, typography, and content strategies using first-principles thinking. Use when researching website design, UI patterns, or competitive design analysis.
development
Audits website usability for UX optimization, covering forms, navigation, validation, and microcopy. Use when reviewing user experience, task completion flows, or interface friction points.