llm-toolkit
AGENTS.md1 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 repository29 other files it gives its agents.
- docxSkill
- pdfSkill
- app-icon-studioSkill
- audit-promptSkill
- bootstrap-goal-projectSkill
- boy-scoutSkill
- brand-namingSkill
- create-engineering-skillSkill
- create-prSkill
- elevenlabsSkill
- firebase-adminSkill
- firebase-hosting-basicsSkill
- gmailSkill
- linearSkill
- nano-banana-app-iconSkill
- new-projectSkill
- plainSkill
- product-researchSkill
- pr-qa-browserSkill
- publish-researchSkill
- readme-best-practicesSkill
- researchSkill
- ripple-checkSkill
- roadmap-pageSkill
- scopeSkill
- searchexaSkill
- slackSkill
- triage-pr-commentsSkill
- write-loopSkill
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 6
Test commands should include transient uv dependencies when the repo does not declare them.
≈ “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
Project Conventions
- This is a skills repository. Files (commands, references, skills, etc.) belong in the project root, NOT in a
.claudefolder. Usecommands/,references/, etc. at the top level. - To install the toolkit, follow the install prompt in the README's Quickstart — there is no installer script. If the user just says "install the toolkit" without specifying, default to project scope with symlinks.
- Source files reference their own scripts and references through
~/.agents/skills/<skill>/...— a harness-neutral stand-in for "wherever this skill is installed". Keep authored files on this convention (never a personal folder path); installations adapt it to the user's harness(es) however fits their system. Exception: hookcommand:lines in Claude frontmatter use~/.claude/...because Claude Code executes them. - Test commands should include transient
uvdependencies when the repo does not declare them. For research-skill tests, use commands likeuv run --with pytest --with typer --with requests pytest skills/research/tests/test_youtube.pyrather than bareuv run pytest ..., becausepytestis not available as a bare executable in this repo's uv environment. - Skills with
disable-model-invocation: truealso needagents/openai.yamlsettingpolicy.allow_implicit_invocation: falseso Codex preserves the explicit-only behavior.
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.