{"id":51551340,"url":"https://github.com/Taiizor/agents-md-cookbook","last_synced_at":"2026-07-12T14:01:15.388Z","repository":{"id":364707339,"uuid":"1268846873","full_name":"Taiizor/agents-md-cookbook","owner":"Taiizor","description":"The tested, tool-agnostic AGENTS.md kit — verified templates, a CI linter, and migrators from .cursorrules/CLAUDE.md/Copilot/Windsurf/Cline/Aider.","archived":false,"fork":false,"pushed_at":"2026-07-08T13:44:33.000Z","size":616,"stargazers_count":4,"open_issues_count":0,"forks_count":0,"subscribers_count":1,"default_branch":"main","last_synced_at":"2026-07-08T15:17:53.500Z","etag":null,"topics":["agents-md","agentsmd","ai","ai-agents","aider","claude","cli","coding-agent","cursor","developer-tools","gemini","github-copilot","linter","llm","windsurf"],"latest_commit_sha":null,"homepage":"https://agents.md","language":"TypeScript","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/Taiizor.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":"CONTRIBUTING.md","funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":"SECURITY.md","support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":null,"dco":null,"cla":null}},"created_at":"2026-06-14T02:20:05.000Z","updated_at":"2026-07-08T13:46:30.000Z","dependencies_parsed_at":null,"dependency_job_id":null,"html_url":"https://github.com/Taiizor/agents-md-cookbook","commit_stats":null,"previous_names":["taiizor/agents-md-cookbook"],"tags_count":2,"template":false,"template_full_name":null,"purl":"pkg:github/Taiizor/agents-md-cookbook","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Taiizor%2Fagents-md-cookbook","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Taiizor%2Fagents-md-cookbook/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Taiizor%2Fagents-md-cookbook/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Taiizor%2Fagents-md-cookbook/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/Taiizor","download_url":"https://codeload.github.com/Taiizor/agents-md-cookbook/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Taiizor%2Fagents-md-cookbook/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":35393398,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-05-26T15:22:16.424Z","status":"online","status_checked_at":"2026-07-12T02:00:06.386Z","response_time":87,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["agents-md","agentsmd","ai","ai-agents","aider","claude","cli","coding-agent","cursor","developer-tools","gemini","github-copilot","linter","llm","windsurf"],"created_at":"2026-07-10T00:00:52.212Z","updated_at":"2026-07-12T14:01:15.372Z","avatar_url":"https://github.com/Taiizor.png","language":"TypeScript","funding_links":[],"categories":["Linting"],"sub_categories":[],"readme":"# agents-md-cookbook\n\n\u003e **The tested, tool-agnostic AGENTS.md kit.**\n\n[![CI](https://github.com/Taiizor/agents-md-cookbook/actions/workflows/ci.yml/badge.svg)](https://github.com/Taiizor/agents-md-cookbook/actions/workflows/ci.yml)\n[![npm: agents-md-lint](https://img.shields.io/npm/v/agents-md-lint?label=agents-md-lint)](https://www.npmjs.com/package/agents-md-lint)\n[![npm: agents-md-migrate](https://img.shields.io/npm/v/agents-md-migrate?label=agents-md-migrate)](https://www.npmjs.com/package/agents-md-migrate)\n[![GitHub Marketplace](https://img.shields.io/badge/Marketplace-agents--md--lint-2ea44f?logo=github)](https://github.com/marketplace/actions/agents-md-lint)\n[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](./LICENSE)\n[![GitHub stars](https://img.shields.io/github/stars/Taiizor/agents-md-cookbook?style=social)](https://github.com/Taiizor/agents-md-cookbook/stargazers)\n\nCopy a battle-tested `AGENTS.md` into your repo, lint it in CI, and migrate your\nold `CLAUDE.md` / `.cursorrules` in one command. Every template is short on\npurpose, leads with runnable commands, and is checked by our own linter.\n\n`AGENTS.md` is a single Markdown file at your repo root that tells coding agents\nhow to build, test, and behave in your project. It is stewarded by the Linux\nFoundation's Agentic AI Foundation, lives in 60,000+ repositories, and is read\nnatively by 24+ tools. One file, every agent.\n\n---\n\n## 30-second value prop\n\n- **Tool-agnostic.** One `AGENTS.md` works across Cursor, Codex, Copilot,\n  Windsurf, Cline, Zed, Jules, Amp, and more. See the full\n  [compatibility matrix](./COMPATIBILITY.md).\n- **Tested, not vibes.** Templates follow evidence from GitHub's 2,500-repo\n  study, ETH Zurich's AGENTbench, and Augment's golden-PR eval — and CI lints\n  every template with our own [`agents-md-lint`](#tooling).\n- **Short by design.** 60-150 lines, commands first, real code snippets,\n  explicit boundaries. No platitudes, no architecture essays, no token bloat.\n- **Migrate in one command.** Turn an existing `CLAUDE.md`, `.cursorrules`, or\n  `GEMINI.md` into a clean `AGENTS.md` with [`agents-md-migrate`](#tooling).\n\n## Quick start\n\n1. **Copy a template** for your stack into your repo root as `AGENTS.md`\n   (filename must be exactly `AGENTS.md` — uppercase `AGENTS`, lowercase `.md`):\n\n   ```bash\n   curl -fsSL \\\n     https://raw.githubusercontent.com/Taiizor/agents-md-cookbook/main/templates/typescript-node/AGENTS.md \\\n     -o AGENTS.md\n   ```\n\n   Browse the full [template index](#templates) for other stacks.\n\n2. **Already have agent instructions?** Migrate them instead of starting over:\n\n   ```bash\n   bunx agents-md-migrate          # or: npx agents-md-migrate\n   ```\n\n   This detects `CLAUDE.md`, `.cursorrules`, `GEMINI.md`, and friends, then\n   produces a single `AGENTS.md`. (Published by the\n   [`agents-md-migrate`](#tooling) package.)\n\n3. **Lint it in CI** so the file stays accurate as your project grows:\n\n   ```yaml\n   # .github/workflows/ci.yml\n   - name: Lint AGENTS.md\n     run: bunx agents-md-lint AGENTS.md   # or: npx agents-md-lint AGENTS.md\n   ```\n\n   Prefer a one-line GitHub Action? Use the bundled [`action.yml`](./action.yml):\n\n   ```yaml\n   # .github/workflows/ci.yml\n   - uses: Taiizor/agents-md-cookbook@v1\n     with:\n       path: AGENTS.md\n   ```\n\n## Compatibility at a glance\n\n| Tool | Reads `AGENTS.md`? |\n| --- | --- |\n| Cursor | NATIVE |\n| OpenAI Codex | NATIVE |\n| GitHub Copilot (coding agent) | NATIVE |\n| Windsurf / Cascade | NATIVE |\n| Cline | NATIVE |\n| Zed | NATIVE (first-match) |\n| Amp | NATIVE |\n| Google Jules | NATIVE |\n| opencode / RooCode | NATIVE |\n| Claude Code | ADAPTER (symlink / `@AGENTS.md`) |\n| Aider | ADAPTER (`--read AGENTS.md`) |\n| Gemini CLI | CONFIG (opt-in) |\n| Copilot (VS Code chat) | CONFIG (experimental) |\n\nFull mechanism, nesting behavior, own-file fallbacks, and sources are in\n**[COMPATIBILITY.md](./COMPATIBILITY.md)** (with a dated last-verified line).\n\n## Templates\n\nEach template is a complete, copy-pasteable `AGENTS.md` for one stack.\n\n| Stack | Path |\n| --- | --- |\n| TypeScript + Node | [`templates/typescript-node/AGENTS.md`](./templates/typescript-node/AGENTS.md) |\n| Python | [`templates/python/AGENTS.md`](./templates/python/AGENTS.md) |\n| Go | [`templates/go/AGENTS.md`](./templates/go/AGENTS.md) |\n| Rust | [`templates/rust/AGENTS.md`](./templates/rust/AGENTS.md) |\n| Java + Spring Boot | [`templates/java-spring/AGENTS.md`](./templates/java-spring/AGENTS.md) |\n| .NET / C# | [`templates/dotnet-csharp/AGENTS.md`](./templates/dotnet-csharp/AGENTS.md) |\n| Next.js | [`templates/nextjs/AGENTS.md`](./templates/nextjs/AGENTS.md) |\n| React + Vite | [`templates/react-vite/AGENTS.md`](./templates/react-vite/AGENTS.md) |\n| Django | [`templates/django/AGENTS.md`](./templates/django/AGENTS.md) |\n| FastAPI | [`templates/fastapi/AGENTS.md`](./templates/fastapi/AGENTS.md) |\n| Ruby on Rails | [`templates/rails/AGENTS.md`](./templates/rails/AGENTS.md) |\n| Monorepo (root) | [`templates/monorepo/AGENTS.md`](./templates/monorepo/AGENTS.md) |\n| Data / ML | [`templates/data-ml/AGENTS.md`](./templates/data-ml/AGENTS.md) |\n| React Native / Expo | [`templates/react-native/AGENTS.md`](./templates/react-native/AGENTS.md) |\n| Minimal starter | [`templates/minimal/AGENTS.md`](./templates/minimal/AGENTS.md) |\n\n## Why AGENTS.md\n\nCoding agents **auto-execute** the commands you list, so a single file that is\naccurate, short, and command-first beats scattered per-tool config. The\nresearch is blunt about what works:\n\n- **Lead with commands and flags.** GitHub's review of 2,500+ repos found the\n  best files put build/test commands early with exact flags, prefer code\n  examples over prose, and pair every boundary with a \"do.\"\n- **Keep it short and non-inferable.** ETH Zurich's AGENTbench found\n  auto-generated `/init` files *reduce* task success ~3% and raise cost\n  ~20-23%; agents follow files faithfully, so every wrong line is executed.\n  Limit content to what an agent cannot infer (custom tooling, unusual builds).\n- **Aim for 100-150 lines.** Augment's golden-PR eval found a 100-150 line\n  sweet spot, numbered workflows (+25% correctness), decision tables, and\n  3-10 line real-code examples (+20% reuse); gains reverse beyond ~300 lines.\n\nRead the full handbook: [anatomy](./docs/anatomy.md) ·\n[best practices](./docs/best-practices.md) ·\n[nesting \u0026 monorepos](./docs/nesting-monorepos.md) ·\n[common mistakes](./docs/common-mistakes.md).\n\n## Tooling\n\nPublished to npm so you can run them with `bunx` or `npx` (no install needed):\n\n- **`agents-md-lint`** — scores an `AGENTS.md` against the evidence-based rules\n  and fails CI on regressions.\n- **`agents-md-migrate`** — converts existing `CLAUDE.md` / `.cursorrules` /\n  `GEMINI.md` into a clean `AGENTS.md`.\n\n\u003e These CLIs live under `packages/` and are released by separate plans; this\n\u003e repo dogfoods `agents-md-lint` on its own templates in CI.\n\n## Contributing\n\nTemplates and matrix updates are the lifeblood of this repo. To add a stack,\nfix a command, or correct a tool's support status, read\n**[CONTRIBUTING.md](./CONTRIBUTING.md)** and open an issue from one of our\n[templates](./.github/ISSUE_TEMPLATE). Every PR is link-checked and\ncontent-linted automatically.\n\n## Credits \u0026 sources\n\n- The `AGENTS.md` standard: \u003chttps://agents.md/\u003e\n- GitHub Blog — *How to write a great agents.md: Lessons from over 2,500\n  repositories* (Matt Nigh, 2025-11-25).\n- ETH Zurich — *Evaluating AGENTS.md* (arXiv:2602.11988, Feb 2026), benchmark\n  AGENTbench.\n- Augment Code — AuggieBench golden-PR evaluation.\n- philschmid — practical AGENTS.md length/structure guidance.\n\n## License\n\n[MIT](./LICENSE) © 2026 Taiizor\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2FTaiizor%2Fagents-md-cookbook","html_url":"https://awesome.ecosyste.ms/projects/github.com%2FTaiizor%2Fagents-md-cookbook","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2FTaiizor%2Fagents-md-cookbook/lists"}