johnnygrecojohnnygreco/skills/AGENTS.md

skills

AGENTS.md1 starsChanged 18 days agoApache-2.0View on GitHub

This repository publishes reusable Agent Skills. Run validation before finishing skill changes: The repository validator accepts the boolean disable-model-invocation extension and checks that invocation policy agrees with agents/openai.yaml for Codex. For skills without extensions, validate one skill directly with the reference validator: If a skill includes scripts, run a representative sample and report the command used.

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 31,700 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 repository10 other files it gives its agents.
- babysit-prSkill
- blast-radiusSkill
- file-prSkill
- full-sendSkill
- full-stopSkill
- leaving-pr-commentsSkill
- panel-reviewSkill
- repo-spring-cleaningSkill
- show-me-your-workSkill
- sol-loopSkill

The file

Show its rules and text

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

15 rules. About 2 are common, 6 are distinctive, and 7 are very specific to this project.

≈2
Common · estimate
Many files like this have a rule of this kind.
≈6
Distinctive · estimate
This author's own choices. Worth a look.
7
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:
Put each skill directly in skills/<skill-name>.
line 7
Very specific:
Keep temporary reports, QA evidence, and other task artifacts in agent-context/working/ (gitignored).
line 8
Very specific:
Do not create auxiliary docs inside individual skill folders.
line 9
Very specific:
Use the skill-creator guidance when available.
line 13
Distinctive:
Keep skill names lowercase with digits and hyphens only.
line 14
Very specific:
Put all trigger guidance in the frontmatter description.
line 16
Very specific:
Keep SKILL.md concise and move detailed material to linked reference files.
line 17
Very specific:
Add agents/openai.yaml when Codex UI metadata, invocation policy, or tool dependencies are useful.
line 18
Common:
Do not commit secrets, private URLs, proprietary snippets, or local absolute paths.
line 19
Distinctive:
Run validation before finishing skill changes:
line 23
Distinctive:
If a skill includes scripts, run a representative sample and report the command used.
line 38
Distinctive:
Keep repository-level docs in the root (README.md, AGENTS.md), not inside individual skill folders.
line 42
Distinctive:
Keep each skill focused on one job.
line 43
Distinctive:
Test any scripts a skill ships before publishing.
line 45
Common:
Do not commit credentials, private URLs, proprietary material, or machine-local paths.
line 46

≈ “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

Repository Instructions

This repository publishes reusable Agent Skills.

Layout

Put each skill directly in skills/<skill-name>.
Keep temporary reports, QA evidence, and other task artifacts in agent-context/working/ (gitignored).
Do not create auxiliary docs inside individual skill folders. A skill folder should contain only SKILL.md and resources the agent may use: agents/, scripts/, references/, and assets/.

Creating Or Editing Skills

Use the skill-creator guidance when available.
Keep skill names lowercase with digits and hyphens only.
Match SKILL.md frontmatter name to the folder name.
Put all trigger guidance in the frontmatter description; the body is loaded only after selection.
Keep SKILL.md concise and move detailed material to linked reference files.
Add agents/openai.yaml when Codex UI metadata, invocation policy, or tool dependencies are useful.
Do not commit secrets, private URLs, proprietary snippets, or local absolute paths.

Validation

Run validation before finishing skill changes:

scripts/validate-skills.py

The repository validator accepts the boolean disable-model-invocation extension and checks that invocation policy agrees with agents/openai.yaml for Codex.

For skills without extensions, validate one skill directly with the reference validator:

uvx --from skills-ref agentskills validate skills/SKILL_NAME

If a skill includes scripts, run a representative sample and report the command used.

Publishing checklist

Keep repository-level docs in the root (README.md, AGENTS.md), not inside individual skill folders.
Keep each skill focused on one job.
Move large or variant-specific details out of SKILL.md into references/.
Test any scripts a skill ships before publishing.
Do not commit credentials, private URLs, proprietary material, or machine-local paths.

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