docs: add ChatGPT mobile-web public judgment - #239

Summary

Adds a bounded Pythia Lotus judgment for the public signed-out ChatGPT mobile-web audit in safal207/LiminalQAengineer PR #106 at exact source head 2407be212e19a393fcd0d8dd33d9fe444aea663b.

Verdict

ALLOW_BOUNDED_DIAGNOSTIC

The source supports:

• a scoped pass for public route, horizontal layout, compact-height composer, primary control geometry, layout stability and event delivery;
• one confirmed mobile-user-agent branch, explicitly not a defect;
• one repeated first-party public-login console diagnostic with no visible login failure or established user/security impact.

Blocked claims

The packet blocks claims that:

• login is broken or unavailable;
• mobile event delivery failed;
• analytics or user data was lost;
• the composer is obstructed;
• duplicate headings or a touch-target violation are proven;
• a security vulnerability exists;
• native-app or authenticated-chat behaviour was assessed.

Deliverables

• docs/lotus-cases/CHATGPT_MOBILE_WEB_PUBLIC_JUDGMENT.md
• examples/lotus-cases/chatgpt-mobile-web-public-judgment-v1.json
• tests/test_chatgpt_mobile_web_public_judgment.py

Validation

python -m pytest tests/test_chatgpt_mobile_web_public_judgment.py

Exact head

cf15c07e7087f399db1b459c4850f5b4261c9b43

Authority boundary

Public audit only. No login, prompt submission, private-conversation access, external report, security claim, deployment, delivery or merge authority.