ratacatratacat/ratacats-skills/AGENTS.md

ratacats-skills

AGENTS.md9 starsChanged 4 months agoMITView on GitHub

Contributor contract for this repository. A skill is only legitimately "in" the repo when it passes every gate below. One skill has exactly one handle. For a skill named <name>, these must be identical strings: Every skill must have: Run bun scripts/sync.ts after any skill change. Run bun scripts/sync.ts --check before committing. Delete skills/<name>/, then run bun scripts/sync.ts. .github/workflows/check.yml runs bun scripts/sync.ts --check on every push and PR.

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 repository33 other files it gives its agents.
- agentic-scaffoldingSkill
- annas-archive-ebooksSkill
- ansi-artSkill
- clarifySkill
- conjecture-cascadeSkill
- crystallizeSkill
- dianalokadaSkill
- duris-game-knowledgeSkill
- duris-group-combatSkill
- ebook-extractorSkill
- failure-modesSkill
- gpt-pro-loopSkill
- graph-researcherSkill
- improve-codebaseSkill
- jevSkill
- julian-writingSkill
- kalshi-prediction-marketSkill
- lemmaSkill
- medium-paywall-bypassSkill
- ml-learning-methodsSkill
- name-reviewSkill
- no-process-pornSkill
- pm-deep-analysisSkill
- pm-market-analysisSkill
- pm-research-methodologiesSkill
- pm-situation-framingSkill
- prediction-drillSkill
- probe-driven-designSkill
- refine-testsSkill
- remove-ai-sinsSkill
- show-usageSkill
- writing-claude-skillsSkill
- x-undocumented-apiSkill

The file

We split this AGENTS.md into the rules it gives an agent

4 rules. About 0 are common, 2 are distinctive, and 2 are very specific to this project.

≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
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

Distinctive:
After any modification, immediately commit and push the task's changes without asking for confirmation.
line 5
Distinctive:
Every skill must have:
line 18
Very specific:
Run bun scripts/sync.ts after any skill change.
line 26
Very specific:
Delete skills/<name>/, then run bun scripts/sync.ts.
line 30

≈ “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.

Keep what fits and drop the rest. Your version, ready for your agents.

Files

AGENTS.md

AGENTS.md — Ratacat's Skills

Contributor contract for this repository. A skill is only legitimately "in" the repo when it passes every gate below.

After any modification, immediately commit and push the task's changes without asking for confirmation.

Naming

One skill has exactly one handle. For a skill named <name>, these must be identical strings:

The directory name skills/<name>/.
name in skills/<name>/SKILL.md YAML frontmatter.
name of the generated entry in .claude-plugin/marketplace.json.
The install handle: /plugin install <name>@ratacats-skills and npx skills add … --skill <name>.

Gates

Every skill must have:

Canonical skill — skills/<name>/SKILL.md with frontmatter name, description, metadata.category, non-empty metadata.keywords, and metadata.blurb (one plain-language sentence; it becomes the root README table row). description is the agent trigger — write it for skill routing, not for humans; blurb is for humans.
README — skills/<name>/README.md: what it does, good fits, install commands, real setup requirements only.
No symlinks — every file under skills/ is real.
Generated index — .claude-plugin/marketplace.json and the root README.md skills table are generated. Never hand-edit generated blocks.
No plugins tree — plugins/ must not exist.

Run bun scripts/sync.ts after any skill change. Run bun scripts/sync.ts --check before committing.

Removing a skill

Delete skills/<name>/, then run bun scripts/sync.ts.

CI

.github/workflows/check.yml runs bun scripts/sync.ts --check on every push and PR.

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.

Check your setup, free