agentleFS
Sign inStart free
robmclartyrobmclarty/fascicle/CLAUDE.md

fascicle

CLAUDE.md4 starsChanged 15 days agoApache-2.0View on GitHub

Claude Code-specific instructions for this repository. AGENTS.md holds the universal contract: pnpm check:all is the source of truth, all conventions live there, all structured output contracts are defined there. Claude Code reads this file and not AGENTS.md, so the line below pulls it in. This file only adds what is Claude-specific.

  • In this repository2 other files it gives its agents.

The file

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

5 rules. About 0 are common, 5 are distinctive, and 0 are 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.

  • 0

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Implement: Prefer small, focused diffs.

    line 16
  2. Distinctive:

    Do not add comments that narrate what the code does ("// loop through users").

    line 28
  3. Distinctive:

    Do not add defensive code for conditions TypeScript's strict mode already rules out.

    line 29
  4. Distinctive:

    Do not apologize in prose for lint or type failures.

    line 30
  5. Distinctive:

    Do not use em dashes in code comments or docs.

    line 31

≈ “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

CLAUDE.md

Claude Code-specific instructions for this repository.

Start here

AGENTS.md holds the universal contract: pnpm check:all is the source of truth, all conventions live there, all structured output contracts are defined there. Claude Code reads this file and not AGENTS.md, so the line below pulls it in.

@AGENTS.md

This file only adds what is Claude-specific.

Workflow

  1. Plan in text first. For any task larger than a typo fix, explain the plan before editing. Reference files with full paths.
  2. Implement. Prefer small, focused diffs. Keep unrelated changes out.
  3. Verify with pnpm check:all. This is non-negotiable. Do not claim done until it exits 0. Use pnpm check (excludes the opt-in slots) for inner loops.
  4. Summarize what changed. One short paragraph, no bullet firehose.

Tool use

  • fallow MCP server is wired up in .mcp.json. Prefer calling fallow tools (analyze, check_changed) during implementation over waiting for the final pnpm check.
  • ast-grep MCP server is also wired up. Use it when editing rules/ or hunting structural patterns that plain grep can't express.
  • Don't shell out to run pnpm check:all repeatedly during tight loops. pnpm check (default, excludes the opt-in slots) is fine for iteration; for even tighter loops use pnpm check --bail --only <relevant checks>, pnpm check --changed, or pnpm exec tsc --noEmit. Run pnpm check:all once at the end: it adds the opt-in mutation (Stryker, the slowest check), the build and packaging gate slots, the learn walk through the course, and the Playwright e2e slot, and skips only the opt-in security audit.

What Claude should avoid

  • Do not add comments that narrate what the code does ("// loop through users"). Comments explain why, not what.
  • Do not add defensive code for conditions TypeScript's strict mode already rules out.
  • Do not apologize in prose for lint or type failures. Fix them.
  • Do not use em dashes in code comments or docs. This is stricter than AGENTS.md, which bans only the paired em dashes around an aside and allows a single one as a pause, and it targets Claude's habit on purpose.

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