agentleFS
eugnmueller-87eugnmueller-87/claude-blueprint/.claude/skills/claude-md/SKILL.md

claude-md

Skill1 starsChanged 5 months agoMITView on GitHub

Audit and tighten this project's CLAUDE.md against reality — confirm the documented commands actually run, keep it under the line cap, and strip stale decisions or duplicated code style. Use when CLAUDE.md has drifted, grown bloated, or after the stack changed. Trigger phrases "update the project memory", "fix CLAUDE.md", "audit the project instructions".

16 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

3 rules. 0 are common, 3 are distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 13 edited versions we read keep these.

  • 3

    Distinctive

    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:

    If over, that's the first thing to fix.

    line 19
  2. Distinctive:

    Verify every command: Run each command in the Commands section once (lint, test, typecheck, build, dev).

    line 21
  3. Distinctive:

    Keep only: real commands, non-obvious architecture, and WHY-decisions.

    line 31

“Common” is measured: kept by at least 6 in 10 of the 13 edited versions of this file we could read. “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

Trigger: "update the project memory" / "fix CLAUDE.md" / "audit the project instructions"

Standing rules: ../_conventions.md.

CLAUDE.md loads every single turn — every line costs tokens forever. Keep it true, tight, and load-bearing. Never document a command you haven't run. Lead with what you changed.

Steps

  1. Read the current CLAUDE.md and count non-blank lines. Hard cap 50, target 25. If over, that's the first thing to fix.
  2. Verify every command. Run each command in the Commands section once (lint, test, typecheck, build, dev). Any that fails or no longer exists → fix the line to match reality or delete it. Never leave a command that doesn't run. Detect the truth from pyproject.toml / package.json, don't guess.
  3. Cut duplication and bloat:
    • Code style rules → belong in rules/code-quality.md, not here. Remove.
    • Path-specific guidance → belongs in a path-scoped rule. Move it.
    • Anything Claude can discover by exploring (file lists, obvious structure) → delete.
    • Keep only: real commands, non-obvious architecture, and WHY-decisions.
  4. Check the Key Decisions section is current. Stale decisions that no longer hold are worse than none — they mislead. Flag any that contradict the current code; confirm with Eugen before rewriting a decision (that's his call).
  5. Report (terse): line count before → after · commands verified (pass/fail) · what you cut or moved · anything flagged for his sign-off. End with the one remaining thing to tighten, if any.

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.