Commit Graph

335 Commits

Author SHA1 Message Date
wuyang6 45792c8fd5 Support runtime guidance tool interrupts 2026-06-23 15:20:59 +08:00
wuyang6 712bccf814 Reject reasoning text in session titles 2026-06-16 15:10:41 +08:00
wuyang6 c8b5b0659d Fix session title refinement after optimistic save 2026-06-16 14:56:58 +08:00
wuyang6 1d38753d08 Stabilize Jupyter terminal wrapper execution 2026-06-15 17:14:54 +08:00
wuyang6 c795ec1b5d Avoid Jupyter terminal job-control noise 2026-06-12 21:38:53 +08:00
wuyang6 c91f65c856 Filter Jupyter terminal wrapper output 2026-06-12 21:32:06 +08:00
wuyang6 7422931290 Auto-run queued session inputs 2026-06-12 20:17:20 +08:00
wuyang6 c3caa4a1e3 Keep optimistic user message during replay 2026-06-12 20:07:42 +08:00
wuyang6 611284b27a Persist theme before hydration 2026-06-12 19:58:16 +08:00
wuyang6 4a268902bf Persist completed run elapsed time 2026-06-12 19:48:45 +08:00
wuyang6 aac6383fe6 Fix runtime guidance transcript replay 2026-06-12 19:41:26 +08:00
wuyang6 9237d6782a Widen sidebar session titles 2026-06-12 19:27:43 +08:00
wuyang6 780f3447b1 Avoid overlapping session title ellipsis 2026-06-12 19:23:47 +08:00
wuyang6 82b41ce89d Reduce session title action spacing 2026-06-12 19:20:47 +08:00
wuyang6 de6849588e Use ellipsis for session titles 2026-06-12 19:17:33 +08:00
wuyang6 c4f9cca2d7 Tighten session list status slot 2026-06-12 18:42:20 +08:00
wuyang6 cdf27e5bd2 Show runtime guidance in transcript 2026-06-12 18:37:04 +08:00
wuyang6 5bcc22622e Show session run badges 2026-06-12 18:28:00 +08:00
wuyang6 4ba8761db2 Add runtime guidance input queue 2026-06-12 18:15:52 +08:00
wuyang6 00a83ba4c3 Cache session replay on client 2026-06-12 17:28:15 +08:00
wuyang6 ddf06c727d Keep append-only session revisions 2026-06-12 16:56:01 +08:00
wuyang6 384e92e961 Prevent stale session overwrite on new tasks 2026-06-12 16:51:17 +08:00
wuyang6 a2f39e7312 Fix session cancel and new task routing 2026-06-12 16:35:38 +08:00
wuyang6 0a3e044c9d Cancel session background processes 2026-06-12 16:24:25 +08:00
wuyang6 953126e1d3 Refactor session runtime persistence 2026-06-12 16:17:58 +08:00
wuyang6 77d360c1e8 docs: add subagent guidance and replay diff skill 2026-06-12 14:39:07 +08:00
wuyang6 c2f08821b9 sanitize admin tool usage stats 2026-06-12 11:46:53 +08:00
wuyang6 dbd279f333 improve admin usage analytics 2026-06-12 11:44:48 +08:00
wuyang6 26e886fb57 allow fast update to restart system services 2026-06-12 11:36:14 +08:00
wuyang6 c4a0b6187b fix deploy service scope detection 2026-06-12 11:32:58 +08:00
wuyang6 40d7c92a1d fix session routing and subagent visibility 2026-06-12 11:06:04 +08:00
wangsenhao a6e7d2260b 修改干预skill 2026-06-10 21:20:15 +08:00
wangsenhao 54bc581cf2 修改label-master 2026-06-08 14:43:44 +08:00
wuyang6 84e792285b Update label master auto service labels 2026-05-29 19:55:22 +08:00
wuyang6 4665b9a632 add lightweight model training skill 2026-05-28 20:06:54 +08:00
hupenglong1 98b3e586de fix: subprocess-based remote sync, pipeline status, skill updates
- Replace thread-based remote state sync with subprocess (sync_remote_state.py)
  to prevent thread pool exhaustion hanging the API
- Fix pipeline showing 'waiting' when a real step is running alongside a gate
- Fix watcher scanner path for linux accounts mode
- Disable label-master semantic review (Step A.5 + §4.5) per user request
- Correct field names: use is_model_correct_dev for accuracy, origin_predict_dev
  for model output (not cleaned_predict)
- Add prohibition against putting test set data in relabel_candidates

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-05-28 14:35:57 +08:00
wuyang6 6364fb437c Normalize product data prompt formatting 2026-05-28 14:31:34 +08:00
wuyang6 2dc941d6d8 Update vehicle control complexity rules 2026-05-27 21:06:40 +08:00
王森浩 1cefbf30c7 Merge branch 'wsh_dev' into 'main'
feat: backend gate fallback + metric_diff docs + submit_sft fix

See merge request wuyang6/zk-data-agent!5
2026-05-27 17:03:27 +08:00
wangsenhao 85ce8c0e69 fix: session resume, config_set schema, metrics chart, subagent logging
- Fix session resume: remove messages.length>1 guard so sessionStorage
  ID is used as resumeSessionId on first message (prevents new session
  creation when user sends "继续")
- Fix config_set tool schema: add missing `items: {}` to array type in
  oneOf (Azure OpenAI strict validation rejects it otherwise)
- Metrics chart: use actual target set filename from trigger text
  instead of generic "target_subset" key
- Add sub-agent skill call logging to agent_runtime.py (full
  prompt/response, no truncation)
- Various metric enrichment and dedup fixes in server.py

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-05-27 17:02:24 +08:00
wuyang6 74f55a421e fix: align model idle timeout with backend budget 2026-05-26 17:21:24 +08:00
wangsenhao 89e77140e7 chore(skill): slim down model-iteration skill files (-820 lines)
SKILL.md: 653→607 lines, program.md: 1757→1187 lines.
Merged duplicate content, replaced verbose explanations with references.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
2026-05-26 17:03:08 +08:00
wuyang6 931cf2d065 fix: isolate auth cookies per deploy 2026-05-26 16:24:11 +08:00
wangsenhao 4dbf59116f docs: simplify complex_dev/complex_base, remove cleaned_predict row
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
2026-05-26 15:49:23 +08:00
wangsenhao b48032e15c feat: backend gate fallback + metric_diff docs + submit_sft fix
- Auto-inject human-review gate when agent asks question but forgets gate entry
- Add complex_dev/complex_base field docs to program.md metric_diff table
- Fix submit_sft.sh minor issues
- jupyter_runtime improvements

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
2026-05-26 15:42:02 +08:00
王森浩 f3ff345656 Merge branch 'wsh_dev' into 'main'
fix: pipeline排序/runDic映射/label-master repeatable/达标优先级

See merge request wuyang6/zk-data-agent!4
2026-05-26 14:12:04 +08:00
wangsenhao a09d4afacd fix: pipeline排序/runDic映射/label-master repeatable/达标优先级
- pipeline items 按 first_ts 自然时间排序,去掉 hardcoded section_order
- augment step-detail 去掉 R{n-1} 偏移,直接用本轮 runDic(与 resolve_run_ids.sh 实际行为对齐)
- 加 BundledSkill.repeatable 字段,label-master 标记 repeatable:true 跳过 duplicate detection
- SKILL.md/program.md: 目标集≥95%最高优先级,未达标前其他集合轻微下降不回滚
- SKILL.md/program.md: 禁止正则脚本替代 label-master 语义复核
- program.md: 每轮 SFT 前必须展示训练数据组成
- resolve_run_ids.sh: 修正注释(SFT_RUNDIC = 当轮 eval workflow id)

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
2026-05-26 14:10:32 +08:00
wuyang6 20c1946a3c fix: stabilize live activity updates 2026-05-26 11:58:13 +08:00
wuyang6 902c9c86c2 fix: run skill git sync as repo owner 2026-05-25 21:25:20 +08:00
王森浩 970d62b539 Merge branch 'wsh_dev' into 'main'
fix: training panel only opens on explicit skill click

See merge request wuyang6/zk-data-agent!3
2026-05-25 19:35:44 +08:00