agentleFS
vingaming1113vingaming1113/github-readme-generator-skill/github-readme-generator/SKILL.md

github-readme-generator

Skill0 starsChanged 4 months agoMITView on GitHub

Generate polished GitHub README.md files for repositories, profile READMEs, libraries, CLIs, apps, tools, and templates. Use when an AI coding agent such as Codex, Claude Code, OpenCode, Cursor, Cline, Aider, Continue, or another agent needs to create or improve a README with clear positioning, installation and usage sections, Shields.io badges, GitHub stats cards, project structure, screenshots/media placeholders, contribution notes, or fast README automation for an existing codebase.

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. wwwzhouhui/skills_collection★ 292same name
    2. zeiynz/claude-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.

  • In this repository2 other files it gives its agents.

The file

Show its rules and text

We split this skill into the rules it gives an agent

14 rules. About 0 are common, 9 are distinctive, and 5 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈9

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 5

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Create high-quality GitHub README files quickly by combining repository inspection, concise project messaging, proven README structure, …

    line 10
  2. Very specific:

    Read references/readme-playbook.md for the repository intelligence pass, README type matrix, section recipes, and quality checklist when …

    line 19
  3. Very specific:

    Use the bundled renderer in scripts/ when a deterministic first draft is useful, then edit the output for repository-specific accuracy and …

    line 22
  4. Distinctive:

    Verify all commands, links, anchors, badge URLs, media URLs, and generated stats URLs before finalizing.

    line 23
  5. Distinctive:

    Use the bundled script to generate a strong starting point from metadata and explicit inputs:

    line 27
  6. Distinctive:

    Prefer concrete commands copied from the repo over generic placeholders.

    line 57
  7. Distinctive:

    Use diagrams, screenshots, terminal recordings, and generated banners only when they help someone understand or evaluate the project faster.

    line 59
  8. Distinctive:

    Keep generated markdown maintainable: plain headings, stable anchors, no excessive HTML, and no hidden dependency on proprietary services.

    line 61
  9. Very specific:

    Read references/badges-and-stats.md when adding or editing Shields.io badges, GitHub Stats Extended cards, profile stats blocks, …

    line 66
  10. Very specific:

    Read references/readme-playbook.md when the README needs stronger positioning, conditional sections, repository-specific storytelling, or …

    line 70
  11. Very specific:

    Read references/resource-catalog.md when choosing resources for templates, profile README inspiration, badges, stats widgets, banners, …

    line 71
  12. Distinctive:

    Run markdown linting or preview when the repo already uses README tooling.

    line 75
  13. Distinctive:

    Test copied shell commands when safe.

    line 76
  14. Distinctive:

    If replacing an existing README, preserve accurate project-specific details and remove stale claims.

    line 78

≈ “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

GitHub README Generator

Overview

Create high-quality GitHub README files quickly by combining repository inspection, concise project messaging, proven README structure, Shields.io badges, visual resources, and GitHub Stats Extended cards.

Workflow

  1. Inspect the target repository before writing. Prefer rg --files, package metadata, config files, existing docs, examples, tests, workflows, screenshots/media, community files, and git remotes.
  2. Identify the README type:
    • Project README: focus on what the project does, why it matters, install, usage, configuration, development, tests, and contributing.
    • Profile README: focus on personal intro, tech stack, featured work, GitHub stats, links, and contact.
    • Organization README: focus on mission, maintained projects, contribution paths, support, and governance.
  3. Read references/readme-playbook.md for the repository intelligence pass, README type matrix, section recipes, and quality checklist when the README needs to be more than a basic scaffold.
  4. Choose a visual system. Use a small, consistent set of badges, cards, screenshots, diagrams, and media; avoid badge walls, huge centered art, and stats blocks that bury the actual project.
  5. Draft a README that is scannable in the first screen: title, one-line value proposition, compact badges, concise overview, and a fast-start path.
  6. Use the bundled renderer in scripts/ when a deterministic first draft is useful, then edit the output for repository-specific accuracy and tone.
  7. Verify all commands, links, anchors, badge URLs, media URLs, and generated stats URLs before finalizing.

Fast Draft Script

Use the bundled script to generate a strong starting point from metadata and explicit inputs:

scripts/render_readme \
  --project /path/to/repo \
  --owner OWNER \
  --repo REPO \
  --name "Project Name" \
  --tagline "Beautiful one-line value proposition" \
  --template auto \
  --screenshot docs/screenshot.png \
  --demo https://example.com \
  --install "npm install" \
  --usage "npm run dev" \
  --output /path/to/repo/README.md

The renderer is implemented in Rust and compiled on demand by scripts/render_readme. It inspects repo files, package metadata, CI, examples, docs, screenshots, Docker files, env examples, licenses, and community files to generate conditional sections. Use --template cli|app|library|docs|auto to steer structure. If the user asked for a profile README, add --profile. If the repository is not public yet, keep Shields.io and stats image URLs only when they will resolve after publication.

Agent Entry Points

  • Codex/OpenAI: use SKILL.md and agents/openai.yaml.
  • Claude Code: start from the repository root CLAUDE.md, which points back to this workflow.
  • OpenCode and other coding agents: start from the repository root AGENTS.md, then load this file and the referenced badge guide as needed.
  • Generic use: if an agent does not support skill discovery, paste or point it to this folder and ask it to follow SKILL.md.

README Quality Bar

  • Make the first sentence explain the outcome, not the implementation detail.
  • Include only badges that communicate useful state: license, CI, release/version, stars, issues, package version, coverage, docs, or platform support.
  • Prefer concrete commands copied from the repo over generic placeholders.
  • Include screenshots or GIF placeholders only when the user can provide assets or the repo has a visual UI.
  • Use diagrams, screenshots, terminal recordings, and generated banners only when they help someone understand or evaluate the project faster.
  • Mention environment variables, secrets, auth scopes, or API keys without exposing values.
  • Keep generated markdown maintainable: plain headings, stable anchors, no excessive HTML, and no hidden dependency on proprietary services.
  • For stats cards, prefer GitHub Stats Extended over legacy GitHub Readme Stats when the user wants maintained dynamic GitHub cards.

Badge And Stats Reference

Read references/badges-and-stats.md when adding or editing Shields.io badges, GitHub Stats Extended cards, profile stats blocks, self-hosted stats notes, or visual README patterns.

Playbook And Resources

  • Read references/readme-playbook.md when the README needs stronger positioning, conditional sections, repository-specific storytelling, or an audit checklist.
  • Read references/resource-catalog.md when choosing resources for templates, profile README inspiration, badges, stats widgets, banners, diagrams, screenshots, media, or README automation. This catalog incorporates categories from vingaming1113/AwesomeREADME.

Final Checks

  • Run markdown linting or preview when the repo already uses README tooling.
  • Test copied shell commands when safe.
  • Confirm every badge and external card URL is URL-encoded.
  • If replacing an existing README, preserve accurate project-specific details and remove stale claims.

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.