agentleFS
Sign inStart free
ClownnvdClownnvd/readme-forge/SKILL.md

readme-forge

Skill1 starsChanged 49 days agoMITView on GitHub

Create or overhaul bold, polished, evidence-backed repository READMEs with strong above-the-fold presentation, accurate setup instructions, useful diagrams, badges, and deterministic quality checks. Use when a user asks for a professional, impressive, visually distinctive, or "badass" README; skip ordinary prose edits or documentation pages that are not repository landing pages.

2 other projects use a version of this.

Adopted by

Most recently taken up by

  • Other versions2 projects use a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 2

    1. tanmayai23/README-Editor-skill★ 2same name
    2. 0xmortuex/claude-code-skills★ 0same name

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

The file

We split this skill into the rules it gives an agent

12 rules. About 0 are common, 12 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈12

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Never invent benchmarks, adoption numbers, compatibility, security claims, sponsors, CI status, install counts, or working commands.

    line 16
  2. Distinctive:

    Do not add remote tracking pixels, hidden analytics, or externally hosted images that expose private repository visits.

    line 17
  3. Distinctive:

    Read the current README, manifests, entry points, tests, CI, release files, license, and public repository metadata.

    line 21
  4. Distinctive:

    Read references/visual-system.md when adding a banner, badges, diagrams, screenshots, or centered HTML.

    line 23
  5. Distinctive:

    Run the deterministic audit:

    line 32
  6. Distinctive:

    Follow references/qa-checklist.md.

    line 38
  7. Distinctive:

    Use specific nouns and verbs.

    line 43
  8. Distinctive:

    Keep badge rows curated. Each badge must answer status, compatibility, distribution, trust, or ownership.

    line 44
  9. Distinctive:

    Use Mermaid only for relationships, sequence, state, or architecture that prose cannot show as clearly.

    line 46
  10. Distinctive:

    Keep quickstart commands copyable and consistent with the actual package manager and runtime.

    line 47
  11. Distinctive:

    Put deep reference material in docs and link it.

    line 49
  12. Distinctive:

    Default to Bold when the user asks for an impressive or "badass" README.

    line 57

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

README Forge

Build a repository front page that earns attention without trading away truth. Visual drama must amplify verified facts, not disguise missing evidence.

Scope and authority

  • Inspect before writing. A request to review a README does not authorize changing it; a request to create or overhaul one does.
  • Preserve project identity, license, commands, package manager, supported versions, and contributor conventions.
  • Never invent benchmarks, adoption numbers, compatibility, security claims, sponsors, CI status, install counts, or working commands.
  • Do not add remote tracking pixels, hidden analytics, or externally hosted images that expose private repository visits.

Workflow

  1. Read the current README, manifests, entry points, tests, CI, release files, license, and public repository metadata. Build a fact sheet: audience, problem, proof, install path, primary action, architecture, status, and constraints.
  2. Choose one repository archetype and read the matching route in references/blueprints.md: Agent Skill, library/SDK, CLI, application, research/demo, or portfolio/showcase.
  3. Read references/visual-system.md when adding a banner, badges, diagrams, screenshots, or centered HTML.
  4. Design the first viewport in this order: identity, one-line value, proof/status badges, primary install or demo action. Do not begin with a table of contents or a wall of text.
  5. Draft only sections that help the target reader decide, install, use, verify, contribute, or assess limitations. Prefer one strong diagram over several decorative ones.

  6. Generate a local SVG banner when it materially improves identity:

    python scripts/generate_banner.py --title "Project Name" --subtitle "One clear promise" --output assets/readme-banner.svg

  7. Run the deterministic audit:

    python scripts/audit_readme.py README.md --root . --strict
  8. Follow references/qa-checklist.md. Render or inspect the README at desktop and narrow widths when the environment supports it. Verify commands rather than merely formatting them.

Writing rules

  • Lead with the result the project creates, then explain the mechanism.
  • Use specific nouns and verbs. Remove inflated claims such as revolutionary, best-in-class, enterprise-ready, or production-ready unless the repository proves them.
  • Keep badge rows curated. Each badge must answer status, compatibility, distribution, trust, or ownership.
  • Give every image meaningful alt text. Keep critical instructions outside images.
  • Use Mermaid only for relationships, sequence, state, or architecture that prose cannot show as clearly.
  • Keep quickstart commands copyable and consistent with the actual package manager and runtime.
  • Label experimental, incomplete, optional, and measured results honestly.
  • Put deep reference material in docs and link it; the README is a decision and onboarding surface.

Visual intensity levels

  • Clean: typography, spacing, concise badges, no custom art.
  • Bold: branded local banner, one diagram, deliberate section rhythm.
  • Showcase: bold treatment plus screenshots or demo media, only when real assets exist.

Default to Bold when the user asks for an impressive or "badass" README. Do not create decorative assets for a tiny internal script unless requested.

Completion evidence

Report the README path, generated assets, commands or links verified, audit result, and any claims that remain unverified. A visually strong README with broken setup instructions is not complete.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free