chore: ai identification in github comments

[NoamGaash]

I'd like to know when comments are made by 🤖

[github-actions]

Preview: https://s3.amazonaws.com/noam-gaash.co.il/30735204470/open-bus/2997/index.html
Preview Storybook: https://s3.amazonaws.com/noam-gaash.co.il/30735204470/open-bus/2997/storybook/index.html
Test Report (if available): 'https://s3.amazonaws.com/noam-gaash.co.il/30735204470/open-bus/04ef3fcdb7291555344e8255fd170ecae3bc284d/test-results/index.html'

[NoamGaash]

@claude fix:

Run npm run lint

> open-bus-map-search@1.5.0 lint
> tsc --noEmit && eslint --max-warnings 0 && stylelint '**/*.{css,scss}' --max-warnings 0 && prettier . -c -u

Checking formatting...
[warn] CLAUDE.md
[warn] README.md
[warn] Code style issues found in 2 files. Run Prettier with --write to fix.

[claude]

Claude finished @NoamGaash's task in 3m 28s —— View job

---

Prettier formatting fixed ✅

🤖 Heads up: this comment was written by an AI agent (Claude), per the very rule this PR adds.

• Gather context (inspect working tree + PR diff)
• Run Prettier on CLAUDE.md and README.md
• Verify npm run lint passes
• Commit and push to ai-identification

Pushed as 1ee205c. Two tiny whitespace fixes — Prettier wants a blank line between a paragraph and the list that follows it:

 This applies to all interactions, including (but not limited to):
+
 - Opening issues

…plus a collapsed double blank line before ## IMPORTANT - AI AGENTS in CLAUDE.md. No content changed.

---

⚠️ Two things worth your attention

1. README.md — the "Star History" section was deleted. This wasn't part of the lint fix, so I left it alone, but it looks accidental. The PR's -2 deletions are exactly this section from origin/main:

-## Star History
-
-[![](https://api.star-history.com/svg?repos=hasadna/...)](...)
+## IMPORTANT - AI AGENTS

I grepped — "Star History" doesn't appear anywhere else in README.md, so it isn't just moved. Looks like the new section overwrote it rather than being appended after it. Fix this →

2. A local-only lint warning you can ignore. On this runner npm run lint also flags .claude-pr/CLAUDE.md. That's a scratch file the @claude Action creates, excluded via .git/info/exclude (which Prettier doesn't read, unlike .gitignore). It won't exist in a normal CI checkout, so it can't fail your build. I verified the real state by running Prettier over the git-tracked files only:

$ git ls-files -z | xargs -0 npx prettier -c -u
Checking formatting...
All matched files use Prettier code style!

tsc --noEmit, eslint --max-warnings 0 and stylelint all passed too (they run first in the && chain and didn't error).

---

Root cause · Prettier requires a blank line separating a paragraph from a following markdown list; the new AI-agents section in both files lacked one (and CLAUDE.md had a stray double blank line).
Evidence · npm run lint reproduced the failure on both files; npx prettier -u --write CLAUDE.md README.md produced the 2-line diff shown above; git ls-files -z | xargs -0 npx prettier -c -u → "All matched files use Prettier code style!"; git diff origin/main...HEAD -- README.md and git show origin/main:README.md for the Star History finding.
Confidence · High for the lint fix and the Star History deletion (both directly observed in the diff).
Couldn't verify · Whether removing "Star History" was intentional — that's your call, so I didn't touch it. I also can't run the CI job myself; the checks on this push will confirm.

_{Unrelated note: my checkout arrived with the AI-agents section already stripped from the working-tree CLAUDE.md. I restored it with git checkout -- CLAUDE.md before formatting, so the commit contains only whitespace changes — your section is intact.}
· ai-identification

[NoamGaash]

I deleted the star graph on purpose because it's stopped working and we don't really need it.

[sonarqubecloud]

Quality Gate passed

Issues
0 New issues
0 Accepted issues

Measures
0 Security Hotspots
0.0% Coverage on New Code
0.0% Duplication on New Code

See analysis details on SonarQube Cloud

[arielvino]

Lets see if it not start spamming code comments with those identifications.
It had a tendency to take such instructions too seriously...