petecurranpetecurran/sorted/scripts/CLAUDE.md

sorted / scripts

CLAUDE.md0 starsChanged 45 days agoMITView on GitHub

Run everything from the repo root. Several of these start or stop servers, or rewrite data, so check the right-hand columns before running one.

In this repository11 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- content/CLAUDE.mdCLAUDE.md
- evals/CLAUDE.mdCLAUDE.md
- seed/CLAUDE.mdCLAUDE.md
- static/CLAUDE.mdCLAUDE.md
- checkSkill
- evalSkill
- rebrandSkill
- resetSkill
- screenshotsSkill

The file

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

3 rules. About 0 are common, 1 is distinctive, and 2 are very specific to this project.

≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.

Common
Distinctive
Very specific

Distinctive:
Run everything from the repo root.
line 3
Very specific:
FT_RESET=1 wipes the database and uploads: Don't use it on a running demo you want to keep.
line 19
Very specific:
screenshots.py changes the data it runs against: Start a fresh copy for it (/screenshots does this) and never point it at the demo on …
line 20

≈ “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

scripts/

Run everything from the repo root. Several of these start or stop servers, or rewrite data, so check the right-hand columns before running one.

Script	What it does	Needs	Changes
`run.sh`	Runs the app in the foreground on `FT_PORT` (8800)	uv; the model only on Apple silicon	With `FT_RESET=1`, the database and the seed data
`start_server.sh`	The same in the background, logging to `data/server.log`	as above	Kills whatever is on the port first
`build_seed.py`	Makes the simulated reports, the cached readings and the phone-sized photos, dated from today	nothing	`seed/demo_incidents.json`, `history.json`, `model_cache.json`, `photos_web/`
`build_kit.py`	Writes GPS into the five demo-kit photos	nothing	`seed/live_demo/*.jpg`; run `build_seed.py` after it
`build_geo.py`	Rebuilds the map layers	raw extracts in `data/raw/land/`, not in git	`seed/*.geojson`
`demo_places.py`	The fixed places the seed and the kit share; a module, not a script
`brand_check.py`	Checks every page at desktop and phone sizes and makes a contact sheet	a running app (`--base`), Google Chrome, `uv run --with playwright`	`data/brand_check/` only
`screenshots.py`	Takes the README's screenshots	a throwaway app with fresh data on another port, Chrome, Playwright	That app's data, `content/brand.json` for a minute (it always puts it back), `docs/screenshots/`
`room.sh`	The live demo's switches: status, pause, close, off, tunnel, password	a running app on 8800	`content/config.json`, `data/council_password.txt`
`tunnel.sh`	Shares the app through a Cloudflare Quick Tunnel	`cloudflared` (installs it with Homebrew)	Nothing, but it puts the app on the internet

The model needs the laptop to itself. Two copies of the 12B model don't fit in 16 GB. Before running evals/run.py, stop the app or restart it with FT_CLASSIFIER=cache.
FT_RESET=1 wipes the database and uploads. Don't use it on a running demo you want to keep.
screenshots.py changes the data it runs against. Start a fresh copy for it (/screenshots does this) and never point it at the demo on 8800.
Don't run tunnel.sh or room.sh tunnel unless asked. Either one makes the laptop reachable from the internet.

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