8 Commits

Author SHA1 Message Date
9aea6c1cd0 docs(ai): converge external wording to AI-assisted + human review, not full-auto
Per the architecture-review redline (EIP3 决策会纪要: AI claims limited to 'AI-assisted, human-reviewed'; never promise 'AI full-auto development'), soften endpoint @Operation summaries, the default task objective, and AI docs that read as committed full-automation. Endpoint paths/method names unchanged (API contract). Added a capability-boundary disclaimer at the top of AI_AGENT_AUTO_DEVELOPMENT.md. 33 platform-ai tests green.

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2026-06-19 16:39:45 +08:00
8bdde45e89 feat: add AI tool access registry 2026-06-17 08:45:56 +08:00
5eb95a4ef4 feat: add typed page schema registry 2026-06-17 08:39:51 +08:00
b05f0e89b5 feat: add release diff rollback plan 2026-06-17 08:33:47 +08:00
ade5e91ac1 feat: add feature package export preview 2026-06-17 08:30:20 +08:00
5f7e226fa0 feat: add CLI env registry and release verification 2026-06-17 08:21:40 +08:00
c5bfbd3ca3 feat: add AI feature package revisions 2026-06-17 07:56:45 +08:00
5bf641838a feat: add EIP4 agent CLI and skills plan 2026-06-17 07:47:22 +08:00