skills
AGENTS.md38 starsChanged 3 months agoMITView on GitHub
Where files are headingNew AGENTS.md files have more rules about keeping secrets out of the code, and fewer about code style.
Of the AGENTS.md files we read, the 31,774 first seen in the last six months, against the older ones: the kind of rule that gained most share, and the kind that lost most. About files like this in general, not this one, and an estimate, since a kind is a pattern in the words.
In this repository31 other files it gives its agents.
- asyncSkill
- code-reviewSkill
- create-planSkill
- design-workflow-graphsSkill
- eli5Skill
- engineering-backend-architectSkill
- expertsSkill
- getnoteSkill
- github-create-prSkill
- github-release-notes-writerSkill
- github-release-pr-writerSkill
- github-renovate-prsSkill
- gitlab-create-mrSkill
- gitlab-mr-inline-commentSkill
- glab-cliSkill
- google-fonts-curatorSkill
- jevSkill
- mac-keychainSkill
- memory-initSkill
- multi-codex-orchestratorSkill
- opentelemetry-go-reviewerSkill
- otel-changelog-pr-verifierSkill
- otel-go-reviewerSkill
- programmer-motivatorSkill
- short-fiction-scaffolderSkill
- skill-converterSkill
- skill-metadata-maintainerSkill
- studySkill
- subagent-orchestratorSkill
- volcengine-ark-image-generatorSkill
- yuqueSkill
The file
We split this AGENTS.md into the rules it gives an agent
1 rule. About 0 are common, 0 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈0
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 19
Use the name field in each SKILL.md frontmatter as the install identifier.
≈ “Common” is an estimate: at least 1 in 5 of the 40,246 AGENTS.md 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.
Files
Repository Guidelines
Project Structure & Module Organization
This repository combines a small Next.js marketplace app with installable skill content. The marketplace now lives under web/: routes, layout, and global styles are in web/src/app, reusable UI components in web/src/components, and parsing/helpers in web/src/lib. Static assets are under web/public/. Documentation lives in docs/. Each installable skill lives in skills/<skill-name>/ and should include a SKILL.md; related references, docs, and agent configs stay beside it. The repository root doubles as a plugin for both Claude Code (.claude-plugin/) and Codex (.codex-plugin/, .agents/plugins/marketplace.json), each exposing the shared skills/ directory; keep the manifests in sync when the plugin identity changes.
Build, Test, and Development Commands
Run marketplace commands from web/. Use cd web && npm install to install dependencies. Run cd web && npm run dev to start the local Next.js server at http://localhost:3000. Run cd web && npm run lint for the primary quality gate with ESLint 9 and Next.js rules. Run cd web && npm run build to verify production output. In restricted environments where Turbopack build workers fail, use cd web && npx next build --webpack as a fallback.
Coding Style & Naming Conventions
Write TypeScript and React with 2-space indentation omitted in favor of the repository’s current style: concise files, semicolons, and single quotes. Keep React components and files in PascalCase, such as SkillCard.tsx; utility modules use lowercase names such as utils.ts. Prefer App Router conventions in web/src/app, and keep skill directory names kebab-case, matching the skill install name where possible.
Testing Guidelines
There is no dedicated automated test suite yet. Treat cd web && npm run lint and cd web && npm run build as required validation before opening a PR. For content changes, manually verify the affected skill page and confirm the SKILL.md frontmatter parses correctly. If you add tests later, place them near the relevant module or in a clearly named tests/ directory and use descriptive names like skills-api.test.ts.
Commit & Pull Request Guidelines
Recent history follows Conventional Commit style, for example feat(subagent-orchestrator): add Codex subagent orchestration skill and chore(deps): update nextjs monorepo to v16.2.1. Keep commit subjects imperative and scoped when useful. PRs should explain the user-visible change, list validation performed (cd web && npm run lint, cd web && npm run build), and link related issues. Include screenshots for UI changes and note any new skill slugs, routes, or metadata expectations.
Content & Configuration Notes
Use the name field in each SKILL.md frontmatter as the install identifier. Keep marketplace URL slugs and install names aligned unless there is a clear reason not to. Avoid introducing required environment variables unless deployment documentation is updated in docs/engineering-guide.md.
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
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.