kinhluankinhluan/skills/AGENTS.md

skills

AGENTS.md6 starsChanged 6 months agoMITView on GitHub

The canonical skill source and platform discovery root is the real skills/ directory. Preserve unrelated work. For any skill change: Do not add SKILL.toon, skill-local README files, machine-specific paths, or skills-lock.json. Verify current technical/research claims against primary sources. For Fieldbook changes:

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 repository75 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- agent-browserSkill
- agent-desktopSkill
- agent-expertise-protocolSkill
- ai-figure-generationSkill
- api-securitySkill
- architecture-decision-recordsSkill
- architecture-reviewSkill
- art-of-war-software-engineeringSkill
- browser-automationSkill
- business-product-leadershipSkill
- c4-level1-contextSkill
- c4-level2-containerSkill
- c4-level3-componentSkill
- c4-level4-codeSkill
- c4-modelSkill
- clean-architectureSkill
- cloud-securitySkill
- code-quality-gateSkill
- collaborative-engineering-agentSkill
- conference-paperSkill
- container-securitySkill
- database-designSkill
- ddd-coreSkill
- ddd-patternsSkill
- ddd-tacticalSkill
- defense-prepSkill
- dependency-managementSkill
- diffusion-release-trackingSkill
- docker-containerizationSkill
- dora-coreSkill
- evolutionary-architectureSkill
- exam-rubricSkill
- experiment-trackingSkill
- federated-learning-dqnSkill
- git-workflowSkill
- golang-developmentSkill
- internal-critiqueSkill
- javascript-typescriptSkill
- journal-q1-polishSkill
- kinhluan-routerSkill
- kubernetes-orchestrationSkill
- merge-request-reviewSkill
- milestone-trackerSkill
- paper-auditSkill
- paper-writingSkill
- pastureSkill
- payment-systemsSkill
- penetration-testingSkill
- phd-proposalSkill
- problem-discoverySkill
- product-analyticsSkill
- product-ux-researchSkill
- progress-reportSkill
- proposal-defenseSkill
- publication-strategySkill
- python-developmentSkill
- research-designSkill
- research-questionSkill
- research-watchSkill
- The first 60 shown.

The file

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

9 rules. About 0 are common, 1 is distinctive, and 8 are very specific to this project.

≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
8
Very specific
Tied to their project's names, files or tools. Probably skip.

Common
Distinctive
Very specific

Very specific:
Keep SKILL.md under 500 lines and link every references/ file directly.
line 7
Very specific:
Put activation context in frontmatter description.
line 8
Very specific:
Update scripts/skill_router.py and regression tests when routing changes.
line 9
Very specific:
Run make sync-manifest, make package, then make check.
line 10
Very specific:
Commit source, skills.json, and every changed deterministic dist/*.skill package together.
line 11
Very specific:
Do not add SKILL.toon, skill-local README files, machine-specific paths, or skills-lock.json.
line 13
Very specific:
Keep skills.json and skills/*/SKILL.md as the only skill-content source.
line 17
Distinctive:
Keep client-side interactions usable without accounts or external data services.
line 19
Very specific:
Run make site-check and keep internal links valid before publishing.
line 20

≈ “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

The canonical skill source and platform discovery root is the real skills/ directory. Preserve unrelated work.

For any skill change:

Keep SKILL.md under 500 lines and link every references/ file directly.
Put activation context in frontmatter description.
Update scripts/skill_router.py and regression tests when routing changes.
Run make sync-manifest, make package, then make check.
Commit source, skills.json, and every changed deterministic dist/*.skill package together.

Do not add SKILL.toon, skill-local README files, machine-specific paths, or skills-lock.json. Verify current technical/research claims against primary sources.

For Fieldbook changes:

Keep skills.json and skills/*/SKILL.md as the only skill-content source.
Preserve the /skills GitHub Pages base configured in site/astro.config.mjs.
Keep client-side interactions usable without accounts or external data services.
Run make site-check and keep internal links valid before publishing.

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