16 Commits

Author SHA1 Message Date
dafang 12a0654eeb chore(eval): remove retired skill experiments 2026-08-03 16:42:46 +08:00
dafang 748d4d3851 feat(cs-issue): add diagnosis-only workflow 2026-08-02 21:35:07 +08:00
dafang 215a0f01f8 test: separate v2 and maintainer eval suites 2026-08-02 16:24:28 +08:00
dafang 4ef7c3fb71 feat(skills): preserve reviewer lineage across re-reviews 2026-08-02 14:50:32 +08:00
dafang 3b80297a08 fix(skills): bound project lesson validation 2026-08-02 14:02:14 +08:00
dafang 83669b3d34 fix: make learning transfer A failures actionable 2026-08-02 12:57:47 +08:00
dafang 70ae347b05 chore: freeze learning transfer target attestation 2026-08-02 11:04:49 +08:00
dafang 652949c08d fix: allow required Seatbelt path metadata 2026-08-02 11:02:37 +08:00
dafang 05c511d42b chore: freeze learning transfer target attestation 2026-08-02 09:27:19 +08:00
dafang 92b12babe3 fix: keep Codex provider auth sources atomic 2026-08-02 09:21:26 +08:00
dafang 54785f75ad fix: harden learning transfer target probes 2026-08-02 08:42:55 +08:00
dafang eed9f55056 feat(eval): add paired project learning transfer
Implements LEARN-3 with frozen paired fixtures, crash-safe journals, target isolation probes, and measured learning-transfer verdicts.

Supersedes local checkpoints 3022df5, 715c831, 9fb5d0f, and fe635f1; those attestations were invalidated and never used for real model runs.
2026-08-02 05:21:37 +08:00
dafang c4fd6eead6 feat(codestable): formalize host-agnostic skill contracts 2026-07-14 21:08:46 +08:00
dafang 2248241634 feat: add risk-tiered feature lanes 2026-07-14 11:37:15 +08:00
dafang 00c78b38a6 release: CodeStable 1.0.3 2026-07-11 10:39:45 +08:00
Wyatt Fang 30fecaae5e release(codestable): ship skill workflow v1.0.2 (#41)
收敛 CodeStable 主入口、runtime preflight、goal driver 与 skill 工程化评测闭环。

- 根 cs 对行动请求同轮直转,咨询与介绍保持非执行
- feature/issue/refactor/epic/docs 按仓库事实恢复,旧 stage skill 保留兼容薄壳
- repo-local runtime 支持版本检测、安全自动同步和显式 refresh-runtime
- 完成 Codex/Claude marketplace 1.0.2、升级文档和回归/评测证据

验证:215 tests passed;package/runtime/diff checks passed;独立 review 与 QA 无 unresolved blocking/important findings。
2026-07-10 15:23:56 +08:00