agentleFS
Sign inStart free
SkillkeelSkillkeel/skillkeel-starter/skills/claude-md-init/SKILL.md

claude-md-init

Skill3 starsChanged 57 days agoMITView on GitHub

Use when a repo has no CLAUDE.md, or the user asks to create/refresh CLAUDE.md or project instructions for Claude Code. Generates a short, verified CLAUDE.md from the repo's real build/test/lint commands and layout, and never invents commands.

1 other project uses a version of this.

Adopted by

Most recently taken up by

  • Other versions1 project uses a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 1

    1. imsovikde/souvik-skills★ 1same name

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • In this repository7 other files it gives its agents.

The file

We split this skill into the rules it gives an agent

6 rules. About 0 are common, 5 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈5

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Write CLAUDE.md with these sections, omitting empty ones:

    line 14
  2. Distinctive:

    Conventions: commit style, branch naming, formatting tool, test placement, error-handling idiom if visible.

    line 18
  3. Distinctive:

    Every command listed was executed or dry-run in this session.

    line 24
  4. Distinctive:

    No prose about "best practices".

    line 25
  5. Distinctive:

    Do not include secrets, internal hostnames, or personal names.

    line 26
  6. Very specific:

    Keep under 100 lines; longer instructions go in docs/ and are linked.

    line 27

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill 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
SKILL.md

claude-md-init

A good CLAUDE.md is short, true, and command-centric. Aim for 40 to 80 lines.

Procedure

  1. Inventory (read, do not guess): ls, cat package.json | jq .scripts, Makefile targets, pyproject.toml / setup.cfg, Cargo.toml, go.mod, docker-compose*.yml, .github/workflows/*.yml (what CI actually runs), README.md headings, .editorconfig, lint configs.
  2. Verify each command you intend to list by running its cheapest, read-only form (npm run / pnpm run to list scripts, make -n test, pytest --collect-only -q | tail -1). Do not run installs or anything that writes to the tree; if a check would modify files, dry-run or skip it and say so. Drop anything that fails.
  3. Detect conventions from git: git log --format=%s -30 (commit style), branch naming, whether PRs are squashed.
  4. Write CLAUDE.md with these sections, omitting empty ones:
    • Project: one sentence, stack, runtime versions (from .nvmrc, .python-version, Dockerfile).
    • Commands: install, dev, test (single test example), lint/format, build, migrate. One line each, copy-pasteable.
    • Layout: 5 to 10 lines mapping top-level dirs to responsibilities.
    • Conventions: commit style, branch naming, formatting tool, test placement, error-handling idiom if visible.
    • Do not: files never to edit by hand (generated, lockfiles, migrations history), commands never to run without asking (deploy, db reset).
    • Verification: the exact command(s) that must pass before claiming done.
  5. Show the file, ask for corrections, then write it. If a CLAUDE.md already exists, propose a diff instead of overwriting.

Rules

  • Every command listed was executed or dry-run in this session. No exceptions.
  • No prose about "best practices"; only facts about this repo.
  • Do not include secrets, internal hostnames, or personal names.
  • Keep under 100 lines; longer instructions go in docs/ and are linked.

Verify

Maintainers: bash evals/run.sh /tmp/skillkeel-evals claude-md-init from the plugin root runs this case and grades it; records and transcripts sit in evals/claude-md-init/. evals/claude-md-init/: fixture: Node monorepo with pnpm scripts + GitHub Actions running pnpm test:ci; expected: CLAUDE.md lists pnpm test:ci (not npm test), notes lockfile as do-not-edit, ≤ 100 lines.

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