Commit graph

9 commits

Author SHA1 Message Date
Kjell Tore Guttormsen
767bc06c51 test(ai-psychosis): extend privacy canary to pattern-phrase leak 2026-05-01 17:56:31 +02:00
Kjell Tore Guttormsen
146cf8ba35 test(ai-psychosis): perf.test.mjs enforces hook timing budget 2026-05-01 17:56:03 +02:00
Kjell Tore Guttormsen
3041c90115 feat(ai-psychosis): /interaction-report adds pushback metrics + reader script 2026-05-01 17:41:30 +02:00
Kjell Tore Guttormsen
b798e68e93 feat(ai-psychosis): SKILL.md cites CC0 Constitution + 5-publication framework 2026-05-01 17:38:57 +02:00
Kjell Tore Guttormsen
79a4249e0b feat(ai-psychosis): persist pushback + domain in sessions.jsonl 2026-05-01 17:30:14 +02:00
Kjell Tore Guttormsen
eca30b4682 feat(ai-psychosis): same-invocation valence-aware pushback detection 2026-05-01 17:28:54 +02:00
Kjell Tore Guttormsen
6b7aee2bf1 feat(ai-psychosis): add 12 pushback + 4 domain regex patterns + cross-check existing 25 2026-05-01 17:10:44 +02:00
Kjell Tore Guttormsen
080f2414ad feat(ai-psychosis): add pushback_count + domain_context state fields 2026-05-01 17:06:09 +02:00
Kjell Tore Guttormsen
297867f847 feat: add ai-psychosis plugin to open marketplace
Meta-awareness tools for healthy AI interaction patterns.
Detects reinforcement loops, scope escalation, and compulsive patterns.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-04-06 20:46:09 +02:00