agentleFS
Sign inStart free
petecurranpetecurran/sorted/content/CLAUDE.md

sorted / content

CLAUDE.md0 starsChanged 45 days agoMITView on GitHub

The rules, words and settings an officer can change without touching code. README.md here is the officer's guide to these files and must stay true: if you change how a rule behaves, change it too.

The file

Create a free account to mount
Files
CLAUDE.md

content/

The rules, words and settings an officer can change without touching code. README.md here is the officer's guide to these files and must stay true: if you change how a rule behaves, change it too.

  • Edits apply on the next page load. common.load_json re-reads a file when its modification time changes, so there's no restart. Screens opened with ?live=1 reload by themselves.
  • A broken file fails quietly. Invalid JSON is logged and the last good copy stays in use, so the edit looks as if it did nothing. The edit hook checks JSON after every change; check by hand with python3 -m json.tool content/<file>.json.
  • Names must match exactly. DEFRA sizes, waste types and land types must be spelt as in common.SIZES, WASTE and LAND, and wards as in seed/wards.geojson. The rules skip a misspelt name without complaint; tests/test_units.py catches it.
  • The order of the rules lives in code. triage.run fixes the precedence (specialist, bulky booking, bin day, a model "no" or "unsure", hold, crew). triage_rules.json holds the lists and the wording. Changing the order means changing triage.py, README.md here and decision 2 together.
  • scripts/room.sh edits config.json with a regular expression. Keep each audience_* switch a quoted string on one line, as "audience_ai": "on", or the script can't find it.
  • Keys starting with _ are notes for people. The app ignores them, so they're the place to explain a value.
  • brand.json follows its own _about note. After changing it, run /rebrand or scripts/brand_check.py.

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