agentleFS
Sign inStart free
kyoo-147kyoo-147/MinimalShip/CLAUDE.md

MinimalShip

CLAUDE.md0 starsChanged 24 days agoMITView on GitHub

This repository is a composable product-development skill bundle. Read AGENTS.md before changing it. Every promoted skill under skills/engineering/ or skills/productivity/ must have: The scripts/check.py checker is the source of truth for this contract. Run it after adding, removing, renaming, or changing a promoted skill. User-invoked skills are explicit commands such as routers, setup, interviewing, planning, and implementation. Model-invoked skills are reusable disciplines such as testing, diagnosis, design, review, and research. Do not make a user-invoked skill depend on another user-invoked…

The file

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

3 rules. About 0 are common, 0 are distinctive, and 3 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.

  • 3

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Every promoted skill under skills/engineering/ or skills/productivity/ must have:

    line 7
  2. Very specific:

    Keep package.json and .claude-plugin/plugin.json versions synchronized.

    line 32
  3. Very specific:

    Use scripts/install.py for Pi installation.

    line 40

≈ “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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
CLAUDE.md

MinimalShip repository conventions

This repository is a composable product-development skill bundle. Read AGENTS.md before changing it.

Promotion contract

Every promoted skill under skills/engineering/ or skills/productivity/ must have:

  • SKILL.md with name and description frontmatter
  • agents/openai.yaml with display metadata
  • matching invocation policy in both files
  • a page under handbook/<bucket>/<name>.md
  • an entry in the bucket README and root README
  • an entry in .claude-plugin/plugin.json

The scripts/check.py checker is the source of truth for this contract. Run it after adding, removing, renaming, or changing a promoted skill.

Invocation

User-invoked skills are explicit commands such as routers, setup, interviewing, planning, and implementation. Model-invoked skills are reusable disciplines such as testing, diagnosis, design, review, and research. Do not make a user-invoked skill depend on another user-invoked skill.

Public content

Write public guidance in original English. Keep product-development scope only. Preserve useful ideas from the supplied MinimalShip v0.3 bundle, including Quick/Deep grilling, output contracts, performance criteria, anti-overengineering boundaries, and truthful status reporting. Do not copy third-party prose or publish the raw sources used for research.

Documentation boundary

docs/, analysis/, and .local/ are private working areas and ignored by Git. Use handbook/ for public human-facing pages. Do not weaken the ignore rule merely to add repository docs.

Release and validation

Keep package.json and .claude-plugin/plugin.json versions synchronized. Run:

python scripts/check.py
bash scripts/list-skills.sh
node scripts/sync-plugin-version.mjs --check

Use scripts/install.py for Pi installation. It must not run downloads, hooks, or arbitrary project code.

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