Commit Graph

1505 Commits

Author SHA1 Message Date
Codex Agent b3e063f38a case-run: 支持 agent mutation prompt 2026-06-06 15:11:31 +08:00
Codex Agent 4e0481977b case-run: 严格提取 agent HWPOD 命令 2026-06-06 14:49:08 +08:00
Codex Agent c2c016b20f case-run: 记录 agent compile trace 2026-06-06 14:37:20 +08:00
Codex Agent 24657acd0c docs: 固化 CaseRun agent compile-only 决策 2026-06-06 14:29:21 +08:00
Lyon 00ddaf68d1 Merge pull request #990 from pikasTech/fix/issue955-v1-trace-contract
fix: 收敛公共 trace result 契约
2026-06-06 14:21:58 +08:00
Codex Agent c58a7fff6c fix: 收敛公共 trace result 契约 2026-06-06 14:20:57 +08:00
Codex Agent 494d3ed3bb fix: add async CaseRun CLI controls 2026-06-06 14:00:18 +08:00
Lyon d5ed5ceed9 Merge pull request #988 from pikasTech/fix/issue955-trace-authority-cleanup
fix: 收敛 Code Agent trace 终态证据路径
2026-06-06 13:57:31 +08:00
Codex Agent 6cae643161 fix: remove legacy trace fallback paths 2026-06-06 13:55:03 +08:00
Lyon ec7e24d4ac Merge pull request #987 from pikasTech/fix/issue986-hwpod-node-command-resolution
fix: stabilize hwpod-node Windows git command resolution
2026-06-06 13:50:37 +08:00
Codex Agent 352da4c436 fix: resolve hwpod node windows git command 2026-06-06 13:49:21 +08:00
Lyon a79c6248ad Merge pull request #985 from pikasTech/fix/issue981-cloud-web-econnreset
fix: keep v0.2 Cloud Web alive on client disconnect
2026-06-06 13:37:37 +08:00
Codex Agent a94b76b94c fix: keep Cloud Web alive on client disconnect 2026-06-06 13:34:34 +08:00
Lyon 28a291996e fix: skip v02 runtime-ready for source-only releases (#984)
Co-authored-by: Codex Agent <codex@hwlab.local>
2026-06-06 13:31:49 +08:00
Lyon e5e376bfcc Merge pull request #983 from pikasTech/fix/issue976-autoeval-summary
fix: keep CaseRun autoEvaluation summaries false
2026-06-06 13:27:30 +08:00
Codex Agent fd5754083a fix: keep CaseRun autoEvaluation summaries false
Keep CaseRun build and collect summaries aligned with evidence.autoEvaluation=false so flow-only runs do not look auto-graded.

Co-Authored-By: Codex <noreply@openai.com>
2026-06-06 13:26:44 +08:00
Lyon 4eb5811360 Merge pull request #980 from pikasTech/fix/issue955-validate-live-cache
fix: 统一 AgentRun trace result 权威路径
2026-06-06 13:10:00 +08:00
Codex Agent 1830df467e fix: unify AgentRun trace result authority 2026-06-06 13:08:23 +08:00
Lyon a97fd6c3e2 Merge pull request #978 from pikasTech/fix/issue976-caserun-agent-stage
fix: 编排 CaseRun Code Agent 流程但不自动评价
2026-06-06 12:55:43 +08:00
Codex Agent 44a59d4f9e fix: orchestrate caserun agent flow without grading 2026-06-06 12:53:31 +08:00
Codex Agent 7f6007d53a docs: clarify CaseRun agent boundary 2026-06-06 12:42:33 +08:00
Codex Agent 98e27f6cb8 docs: require main worktree for docs and CLI changes 2026-06-06 12:33:17 +08:00
Lyon dfdb920848 Merge pull request #975 from pikasTech/fix/issue955-repair-ambiguous-top-level
fix: repair ambiguous top-level trace snapshots
2026-06-06 12:15:50 +08:00
Codex Agent f4b33b5cbb fix: repair ambiguous top-level trace snapshots 2026-06-06 12:15:06 +08:00
Lyon a210728045 Merge pull request #974 from pikasTech/fix/issue955-validate-trace-evidence
fix: validate trace result provenance
2026-06-06 12:04:29 +08:00
Codex Agent c80493f240 fix: validate trace result provenance 2026-06-06 12:03:29 +08:00
Lyon 3d073fee7e Merge pull request #973 from pikasTech/fix/issue955-preserve-latest-trace
fix: preserve latest trace during historical repair
2026-06-06 11:41:03 +08:00
Codex Agent aa06af6a91 fix: preserve latest trace during historical repair 2026-06-06 11:40:21 +08:00
Lyon 62aeb9411d Merge pull request #972 from pikasTech/fix/issue955-historical-trace-index
fix: index historical AgentRun traces
2026-06-06 11:22:30 +08:00
Codex Agent 7c6df8d9b4 fix: index historical AgentRun traces 2026-06-06 11:21:00 +08:00
Codex Agent 4a4aa1bc29 feat: isolate CaseRun subject worktrees 2026-06-06 11:09:08 +08:00
Lyon 8507956d8f Merge pull request #971 from pikasTech/fix/issue955-web-renderer-ci-dep
fix: include trace renderer in cloud web rollout plan
2026-06-06 10:48:34 +08:00
Codex Agent d300428246 fix: include trace renderer in cloud web rollout plan 2026-06-06 10:46:44 +08:00
Lyon 11897b700b Merge pull request #970 from pikasTech/fix/issue955-trace-replay-final
fix: stabilize AgentRun trace replay and final rows
2026-06-06 10:35:06 +08:00
Codex dd0121c93c fix: stabilize AgentRun trace replay and final rows 2026-06-06 10:33:26 +08:00
Lyon e2c33ae3e6 Merge pull request #969 from pikasTech/fix/issue968-caserun-registry
fix: migrate CaseRun registry to hwlab-case-registry
2026-06-06 10:12:48 +08:00
Codex Agent 2a651d11d3 fix: migrate caserun registry to hwlab-case-registry 2026-06-06 10:11:46 +08:00
Lyon 6afdb83e01 Merge pull request #967 from pikasTech/docs/caserun-registry-rename-20260606
docs: rename CaseRun registry repo
2026-06-06 09:56:58 +08:00
Codex Agent 7ddfffcd02 docs: rename caserun registry repo 2026-06-06 09:46:56 +08:00
Lyon 3293f901c6 Merge pull request #966 from pikasTech/docs/caserun-source-repo-boundary-20260606
docs: define CaseRun source repo boundary
2026-06-06 09:43:15 +08:00
Codex Agent 545b1dc892 docs: define caserun source repo boundary 2026-06-06 09:34:00 +08:00
Codex Agent 346be42b8d docs: record hwpod caserun current state 2026-06-06 09:01:01 +08:00
Lyon bfe7c5eab7 Merge pull request #965 from pikasTech/fix/hwpod-spec-discovery-960
fix(v0.2): 发现并预装可用 hwpod-spec
2026-06-06 01:53:49 +08:00
Codex Agent 42b53e5795 fix: discover preinstalled hwpod specs 2026-06-06 01:49:59 +08:00
Lyon 459f2960b3 Merge pull request #964 from pikasTech/fix/issue959-markdown-fidelity
fix: preserve structured markdown final response
2026-06-06 01:38:43 +08:00
Codex Agent c7e210cf96 fix: preserve structured markdown final response 2026-06-06 01:37:44 +08:00
Lyon 711a3ec37a Merge pull request #963 from pikasTech/fix/issue959-natural-markdown-v2
fix: prefer terminal markdown final response
2026-06-06 01:24:25 +08:00
Codex Agent 8f7bc79d1e fix: prefer terminal markdown final response 2026-06-06 01:23:18 +08:00
Lyon eb9267a31f Merge pull request #962 from pikasTech/fix/issue955-stable-replay-timer
fix: keep compact trace replay timers stable
2026-06-06 01:05:45 +08:00
Codex Agent 02bd350f46 fix: keep compact trace replay timers stable 2026-06-06 01:04:39 +08:00