GitHubClawToolkit
AGENTS.md254 starsChanged 7 months agoMITView on GitHub
Use ESM JavaScript for Worker and skill scripts. Keep modules small and purpose-named, such as config.js, line.js, or resolve-worker-name.mjs. Use kebab-case directories and UPPER_SNAKE_CASE environment variables. Worker tests live in workers/line-bot/test/ and use Node’s node --test runner. Name tests with the .test.js suffix, for example config.test.js. Add tests for parsing, configuration, request handling, and branch/name normalization when changing Worker behavior. Skills rely on build checks or script-level manual verification unless they define tests. Use Conventional Commits 1.0.0: <type>[optional scope]: <description>,…
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 repository20 other files it gives its agents.
- best-practices-for-agent-skillsSkill
- githubclaw-json-generatorSkill
- pr-creatorSkill
- skill-creatorSkill
- agent-browserSkill
- felo-searchSkill
- felo-slidesSkill
- felo-superAgentSkill
- felo-web-fetchSkill
- felo-x-searchSkill
- find-skillsSkill
- gemini-audio-transcriberSkill
- gemini-deep-researcherSkill
- gemini-image-describerSkill
- gemini-lyria-3Skill
- gemini-nanobananaSkill
- gemini-summarySkill
- google-stitchSkill
- skill-creatorSkill
- telegram-notifySkill
The file
We split this AGENTS.md into the rules it gives an agent
2 rules. About 0 are common, 0 are distinctive, and 2 are 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.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- could be automatic: a setting could enforce it
- Very specific: line 24
Use ESM JavaScript for Worker and skill scripts.
- Very specific: line 50
Never commit API keys, LINE tokens, GitHub tokens, .dev.vars, or generated dependency directories.
≈ “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
skills/: standalone Agent Skills withSKILL.md,README.md, andgithubclaw.json; addscripts/,src/,references/, orassets/only when needed.templates/: reusable agent or system templates, usually withAGENTS.mdorSYSTEM.mdplusgithubclaw.json.actions/: composite GitHub Actions and reusable workflow assets. Each action hasaction.ymlandREADME.md.workers/line-bot/: Bun/Node Cloudflare Worker source, tests, scripts, migrations, and deployment config.
Build, Test, and Development Commands
cd workers/line-bot && bun install: install Worker dependencies.cd workers/line-bot && bun run build: bundlesrc/index.js.cd workers/line-bot && bun run test: run Node’s built-in test suite.cd workers/line-bot && WORKER_NAME=your-worker-name bun run dev: start Wrangler development mode.cd skills/<skill-name> && bun run build: rebuild skills that providesrc/and apackage.json.
Coding Style & Naming Conventions
Use ESM JavaScript for Worker and skill scripts. Keep modules small and purpose-named, such as config.js, line.js, or resolve-worker-name.mjs. Use kebab-case directories and UPPER_SNAKE_CASE environment variables.
Testing Guidelines
Worker tests live in workers/line-bot/test/ and use Node’s node --test runner. Name tests with the .test.js suffix, for example config.test.js. Add tests for parsing, configuration, request handling, and branch/name normalization when changing Worker behavior. Skills rely on build checks or script-level manual verification unless they define tests.
Commit & Pull Request Guidelines
Use Conventional Commits 1.0.0: <type>[optional scope]: <description>, with optional body and footers. Examples: feat(error-handler-action): ..., fix(workflow): ..., docs: ..., and chore(templates): .... Avoid literal string \n in commit messages; use actual newline characters. Do not create message bodies with repeated git commit -m parameters. Write the full message to a temporary file and run git commit -F /tmp/commit-message.txt.
Pull requests should describe changed paths, link related issues, list validation commands, and attach screenshots or sample outputs for user-facing content.
Action Release Reminder
Reusable workflow consumers are pinned to the v1 tag. If any file under actions/ changes, explicitly remind the maintainer to decide whether to republish or move v1 to the latest compatible commit. Do not move tags unless asked.
Security & Configuration Tips
Never commit API keys, LINE tokens, GitHub tokens, .dev.vars, or generated dependency directories. For workers/line-bot/, copy expected variables from env.example, then use WORKER_NAME=your-worker-name bun run set to upload secrets through Wrangler.
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.