agentleFS
Sign inStart free
RohiRIKRohiRIK/skills/AGENTS.md

skills

AGENTS.md1 starsChanged 30 days agoMITView on GitHub

Personal library of reusable Agent Skills. Each subfolder under skills/ is one skill. Help add and maintain skills correctly. Apply the rules below on every skill-creation or review task. rules/system.md is the source of truth — skill structure, the tier model, frontmatter, composition, state, and telemetry. Read it before creating or wiring skills.

The file

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

1 rule. About 0 are common, 1 is distinctive, and 0 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.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Every skill needs a row in README.md.

    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

Skills — personal Agent Skills library

Personal library of reusable Agent Skills. Each subfolder under skills/ is one skill.

Your role

Help add and maintain skills correctly. Apply the rules below on every skill-creation or review task.

How the system works → rules/system.md

rules/system.md is the source of truth — skill structure, the tier model, frontmatter, composition, state, and telemetry. Read it before creating or wiring skills.

Rules (essentials)

  • Each skill is a folder under skills/ with a SKILL.md carrying name, description (WHAT + WHEN, ≤30 words), category (workflow·reference·delegation·meta·visual·prompting·quality), and effort (low·medium·high).
  • Folder + file names are TitleCase (mirror ~/.claude/skills); the name: field drives activation.
  • Flat structure, 2 levels max; only Workflows/ and Tools/ subdirs; context files in the skill root; SKILL.md ≤ 50 lines.
  • Host portability: name skills (not slash commands) and capability-check host-native tools with an inline fallback.
  • Word skills via the Prompting skill. Reserve CRITICAL/MUST/NEVER for genuine safety gates.
  • Every skill needs a row in README.md.

Skill composition (always-on map)

Primitives:  Verify (quality gate) · Reflect (self-eval)
Drivers:     Iterate → Verify + Reflect each pass (wake primitive if the host has one, inline passes if not)
             Orchestrate → Decompose→RunLayer→MergeQueue; delegates to Agy/OpenCode/Pi; gates with Verify
             Iterate:RunLenses → feeds Spec / Orchestrate
             Research → fans out to Agy/OpenCode/Pi
Meta:        SkillForge (audit + instrument the fleet) · CreateSkill (build one skill) · ModelUpgrade (re-tune for any model, current or new)
State:       .agent-state.md  ·  Telemetry: ~/.claude/state/execution.jsonl

Where to go

  • Full system spec → rules/system.md
  • Repo layout + Claude Code specifics → CLAUDE.md
  • Install a skill into an AI tool → INSTALL-AI.md

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