t4sht4sh/skills4sh/AGENTS.md

skills4sh

AGENTS.md5 starsChanged 46 days agoMITView on GitHub

This is a mono-repo of agent skills. Each skill lives in skills/<skill-name>/ with a SKILL.md file. Always copy, never symlink. Use docs/RELEASE_PROCESS.md as the canonical release and npm publishing runbook. Key invariant: only create a vX.Y.Z tag after the PR containing package.json version X.Y.Z has been merged to main. The GitHub release for that tag triggers npm Trusted Publishing and checks out exactly that tag. Planning, analysis, and testing commits do not require version bumps. Defer bumps until the merge…

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 repository9 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- agent-memorySkill
- code-to-figmaSkill
- discord-harvestSkill
- eleventy-nunjucksSkill
- figma-to-codeSkill
- init-rulebookSkill
- localhost-screenshotsSkill
- skill-architectSkill

The file

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

5 rules. About 0 are common, 2 are distinctive, and 3 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.
3
Very specific
Tied to their project's names, files or tools. Probably skip.

Common
Distinctive
Very specific

Very specific:
Read .agent-memory/index.yaml to discover project memory when .agent-memory/ exists.
line 8
Distinctive:
Check for older installations (remove symlinks silently.
line 16
Distinctive:
Always copy, never symlink.
line 20
Very specific:
Use docs/RELEASE_PROCESS.md as the canonical release and npm publishing runbook.
line 24
Very specific:
When creating, reviewing, or improving skills in this repository, use docs/SKILL_AUTHORING_STANDARD.md as the canonical rubric and …
line 37

≈ “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

Agent Instructions

This is a mono-repo of agent skills. Each skill lives in skills/<skill-name>/ with a SKILL.md file.

First Steps

Read this file.
Read .agent-memory/index.yaml to discover project memory when .agent-memory/ exists.
Load relevant memory files for the task before changing skill content, release metadata, gates, or repo conventions.

When asked to install a skill

Ask the user where to install:
- Global: the host default (see README installer table: ~/.claude/skills/, ~/.cursor/skills/, ~/.grok/skills/, ~/.openclaw/skills/, ~/.agents/skills/, or --dest)
- Project: ./.agents/skills/<skill-name>/, or the project-local equivalent (.claude/skills/, .grok/skills/, workspace skills/ for OpenClaw)
Check for older installations (remove symlinks silently; ask before overwriting directories).
Copy the full skill folder contents (SKILL.md, LICENSE, references/, and any assets/) to the destination.
Show: Installation complete! The skill will now be loaded from <path>. Refresh or restart your session for changes to take effect.

Always copy, never symlink.

Release and npm publish rules

Use docs/RELEASE_PROCESS.md as the canonical release and npm publishing runbook.

Key invariant: only create a vX.Y.Z tag after the PR containing package.json version X.Y.Z has been merged to main. The GitHub release for that tag triggers npm Trusted Publishing and checks out exactly that tag.

Draft skill work

Planning, analysis, and testing commits do not require version bumps. Defer bumps until the merge decision, then synchronize the final versions before merge. Keep content hashes and checks current throughout; follow the authoring standard.

Skill authoring standard

When creating, reviewing, or improving skills in this repository, use docs/SKILL_AUTHORING_STANDARD.md as the canonical rubric and skills/skill-architect/ as the in-house planning/review skill.

In summary:

skill-architect is the operational planning/review skill for this repo's skill work.
Skill Development supplies structural patterns that skill-architect bridges into the local rubric.
writing-skills contributes validation discipline through skill-architect and the local standard.
Anthropic/OpenAI skill-creator patterns inform evals and compatibility; they are inputs, not the repo's primary rubric.

Available skills

Skill	Path
agent-memory	`skills/agent-memory/`
code-to-figma	`skills/code-to-figma/`
discord-harvest	`skills/discord-harvest/`
eleventy-nunjucks	`skills/eleventy-nunjucks/`
figma-to-code	`skills/figma-to-code/`
init-rulebook	`skills/init-rulebook/`
localhost-screenshots	`skills/localhost-screenshots/`
skill-architect	`skills/skill-architect/`

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