shippable-skills
CLAUDE.md7 starsChanged 22 days agoMITView on GitHub
Skills are organized into bucket folders under skills/: Every skill in engineering/ or productivity/ (the promoted buckets) must have a reference in the top-level README.md and an entry in .claude-plugin/plugin.json's skills array. Skills in misc/, in-progress/, and deprecated/ must not appear in either. .claude-plugin/plugin.json's version tracks package.json's version: bump both together on release. Each promoted skill must have a human README.md beside its SKILL.md. That folder URL is the page you share. Catalog entries in the top-level README.md and in…
In this repository8 other files it gives its agents.
- bump-releaseSkill
- exploreSkill
- fix-testsSkill
- readme-pitchSkill
- product-spineSkill
- session-handoffSkill
- session-primerSkill
- session-retroSkill
The file
We split this CLAUDE.md into the rules it gives an agent
4 rules. About 0 are common, 0 are distinctive, and 4 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈0
Distinctive · estimate
This author's own choices. Worth a look.
4
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 11
Every skill in engineering/ or productivity/ (the promoted buckets) must have a reference in the top-level README.md and an entry in …
- Very specific: line 21
Every SKILL.md is either user-invoked (disable-model-invocation: true plus policy.allow_implicit_invocation: false in agents/openai.yaml) …
- Very specific: line 27
Keep SKILL.md under 500 lines.
- Very specific: line 31
No em-dashes anywhere in this repo's prose (SKILL.md files, docs, README.md, CLAUDE.md, comments).
≈ “Common” is an estimate: at least 1 in 5 of the 26,980 CLAUDE.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.
Files
Shippable Skills
Skills are organized into bucket folders under skills/:
engineering/: daily code work (bump-release, fix-tests, explore)productivity/: daily non-code workflow toolsmisc/: kept around but rarely used, not promotedin-progress/: beta, not shipped in the plugindeprecated/: no longer used
Every skill in engineering/ or productivity/ (the promoted buckets) must have a reference in the top-level README.md and an entry in .claude-plugin/plugin.json's skills array. Skills in misc/, in-progress/, and deprecated/ must not appear in either.
.claude-plugin/plugin.json's version tracks package.json's version: bump both together on release.
Each promoted skill must have a human README.md beside its SKILL.md. That folder URL is the page you share. Catalog entries in the top-level README.md and in each bucket README.md link the skill name to the skill directory, not to SKILL.md.
Promoted bucket READMEs and the top-level README group entries into User-invoked and Model-invoked.
Skills in engineering/ and productivity/ also have a human-facing docs page at docs/<bucket>/<skill>.md. When you add, rename, or change the behaviour of a promoted skill, create or re-sync its docs page. Four sections make a page finished: What it does, When to reach for it, Common questions, and It's working if. The full install (npx plus the Claude Code plugin path) lives only in the top-level README.md. A skill-folder README.md may repeat the one-line npx skills add command and link back to the collection README. Docs pages do not host that dual install.
Every SKILL.md is either user-invoked (disable-model-invocation: true plus policy.allow_implicit_invocation: false in agents/openai.yaml) or model-invoked (omit both). Keep the two harnesses in sync: a skill is user-invoked in both or neither.
A user-invoked skill may invoke model-invoked skills, but never another user-invoked one. Operative instructions name the Skill tool (Call the Skill tool with "name"), not a /slash that assumes one host.
Each skill carries an agents/openai.yaml beside its SKILL.md (Codex picker metadata). Shared reference docs live inside the skill that owns them (references/), one level deep from SKILL.md.
Keep SKILL.md under 500 lines. The description is human-facing for user-invoked skills (no "Use when the user says..." trigger lists) and model-facing for model-invoked skills (rich trigger phrasing).
This repo must stay host-neutral. Do not depend on a host plugin-root env var, a single plugin PATH, or one vendor's tool names. Discover skills from the common project and home directories; map host tool aliases when tallying.
No em-dashes anywhere in this repo's prose (SKILL.md files, docs, README.md, CLAUDE.md, comments). Where a sentence reaches for one, rewrite it instead with a comma, colon, period, parentheses, or a conjunction.
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
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.