Kevin 99543d04c6 feat(eval): internal-eval stack, judge fixes, and eval web overhaul ...

- Merge internal-eval into development.sh (single Celery/infra); internal-eval.sh
  wraps with LIFE_ECHO_WITH_INTERNAL_EVAL; EVAL_ATTACH_ONLY for attaching 8001
  when :8000 is already up; document in api/docs/internal-eval.md.
- Evaluation: transcript_for_judge, judge error surfacing, rubric/schema tweaks,
  execution_service and router updates; tests for judge and composite eval.
- Memory: ingest nested transaction for embedding/enrichment rollback safety.
- Conversation WS: logger.exception for pipeline errors (avoid loguru KeyError).
- app-eval-web: Playground saved replays, dialogue turns helper, hash user_id
  for Memoir; Memoir chapter baseline↔DB row compare with title heuristics;
  Stories page (#memoir-stories); Markdown + copy buttons; toolbar/panel UI;
  react-markdown; development proxy and fixture updates.

2026-04-07 17:18:47 +08:00

..

evaluation

feat/ eval

2026-04-06 23:19:20 +08:00

evaluation_calibration

feat(eval): internal-eval stack, judge fixes, and eval web overhaul

2026-04-07 17:18:47 +08:00

fixtures

refactor(chat): AI-native prompts, remove interview heuristics

2026-04-06 22:23:46 +08:00

user_exports

feat(eval): internal-eval stack, judge fixes, and eval web overhaul

2026-04-07 17:18:47 +08:00

test_alembic_revision_ids.py

fix/ 修复后端dockerworkflow会跑两次alembic的问题

2026-04-01 16:29:18 +08:00

test_background_runner.py

feat(memoir): 回忆录分段两阶段管线（Phase1 分类 / Phase2 叙事）与配置、测试

2026-04-02 16:37:14 +08:00

test_background_voice.py

refactor(chat): AI-native prompts, remove interview heuristics

2026-04-06 22:23:46 +08:00

test_chapter_read_view.py

feat(api): 叙事 prompt、职业上下文、读路径章节、WS 解耦与错误脱敏

2026-04-01 11:55:52 +08:00

test_chat_input_normalize.py

feat(api): 访谈路径轻量门控、Memoir Phase1 批处理与叙事/记忆管线加固

2026-04-03 10:12:59 +08:00

test_chat_stage_detection_gates.py

refactor(chat): AI-native prompts, remove interview heuristics

2026-04-06 22:23:46 +08:00

test_chat_stage_detection.py

feat(api+app): 对话阶段化、回忆录流水线与客户端会话体验

2026-03-26 12:13:36 +08:00

test_classification_fragment.py

feat(api): 访谈人格/回复长度策略、口述归一、背景语气与输入净稿全链路

2026-03-31 23:55:26 +08:00

test_eval_composite.py

feat(eval): internal-eval stack, judge fixes, and eval web overhaul

2026-04-07 17:18:47 +08:00

test_experience_regressions.py

refactor(chat): AI-native prompts, remove interview heuristics

2026-04-06 22:23:46 +08:00

test_fidelity_gate.py

feat(api): 统一 LLM JSON 调用层 llm_json_call，按域 Schema 迁移 chat/memoir agents

2026-04-03 13:44:11 +08:00

test_http_contract_errors.py

feat(api): 叙事 prompt、职业上下文、读路径章节、WS 解耦与错误脱敏

2026-04-01 11:55:52 +08:00

test_image_prompt_policy.py

refactor(agents): 抽取阶段常量与对话上下文；快档 LLM；图片 prompt 可禁止回退

2026-04-02 12:00:00 +08:00

test_infra_regressions.py

feat(api): 拆分章节物化与 Story 后处理，并加固 Redis 锁与腾讯 ASR

2026-03-30 11:53:04 +08:00

test_interview_prompts.py

refactor(chat): AI-native prompts, remove interview heuristics

2026-04-06 22:23:46 +08:00

test_json_and_memory_utils.py

refactor(agents): 抽取阶段常量与对话上下文；快档 LLM；图片 prompt 可禁止回退

2026-04-02 12:00:00 +08:00

test_judge_schemas.py

feat(eval): internal-eval stack, judge fixes, and eval web overhaul

2026-04-07 17:18:47 +08:00

test_judge_service.py

feat(eval): internal-eval stack, judge fixes, and eval web overhaul

2026-04-07 17:18:47 +08:00

test_llm_json_call.py

feat(api): 统一 LLM JSON 调用层 llm_json_call，按域 Schema 迁移 chat/memoir agents

2026-04-03 13:44:11 +08:00

test_memoir_idempotency.py

feat(api): 访谈路径轻量门控、Memoir Phase1 批处理与叙事/记忆管线加固

2026-04-03 10:12:59 +08:00

test_memoir_skip_story.py

feat(api): 访谈路径轻量门控、Memoir Phase1 批处理与叙事/记忆管线加固

2026-04-03 10:12:59 +08:00

test_memoir_two_phase.py

feat(memoir): 回忆录分段两阶段管线（Phase1 分类 / Phase2 叙事）与配置、测试

2026-04-02 16:37:14 +08:00

test_memory_compaction.py

聊天和回忆录证据检索都走 pgvector，去掉 Postgres FTS/content_tsv，新迁移删掉 content_tsv 列（部署要先 alembic upgrade）。

2026-04-03 11:43:16 +08:00

test_memory_consistency_rules.py

feat(api): 访谈路径轻量门控、Memoir Phase1 批处理与叙事/记忆管线加固

2026-04-03 10:12:59 +08:00

test_memory_evidence.py

聊天和回忆录证据检索都走 pgvector，去掉 Postgres FTS/content_tsv，新迁移删掉 content_tsv 列（部署要先 alembic upgrade）。

2026-04-03 11:43:16 +08:00

test_narrative_boundary_regressions.py

feat(api): 访谈路径轻量门控、Memoir Phase1 批处理与叙事/记忆管线加固

2026-04-03 10:12:59 +08:00

test_narrative_pipeline.py

feat(memoir): 路由阶段不要求标题，按正文字数门闸延迟 LLM 标题

2026-04-02 14:38:40 +08:00

test_occupation_context.py

feat(api): 叙事 prompt、职业上下文、读路径章节、WS 解耦与错误脱敏

2026-04-01 11:55:52 +08:00

test_oral_normalize.py

feat(api): 叙事 prompt、职业上下文、读路径章节、WS 解耦与错误脱敏

2026-04-01 11:55:52 +08:00

test_recompose_retry_policy.py

feat(api): 访谈路径轻量门控、Memoir Phase1 批处理与叙事/记忆管线加固

2026-04-03 10:12:59 +08:00

test_reply_segments.py

fix/ 修复AI聊天时回复markdown导致聊天气泡布局问题

2026-04-03 14:06:55 +08:00

test_stage_slot_registry.py

feat(api): 统一 LLM JSON 调用层 llm_json_call，按域 Schema 迁移 chat/memoir agents

2026-04-03 13:44:11 +08:00

test_stage_validation.py

feat(api): 统一 LLM JSON 调用层 llm_json_call，按域 Schema 迁移 chat/memoir agents

2026-04-03 13:44:11 +08:00

test_state_service_batch_stage_policy.py

feat(memoir): 回忆录分段两阶段管线（Phase1 分类 / Phase2 叙事）与配置、测试

2026-04-02 16:37:14 +08:00

test_story_route_oral_invariant.py

feat(api): 访谈人格/回复长度策略、口述归一、背景语气与输入净稿全链路

2026-03-31 23:55:26 +08:00

test_story_route_payload.py

fix(memoir): 改善 story 合并决策，少生碎片篇

2026-04-03 11:02:05 +08:00

test_story_route_prompts_and_behavior.py

feat(api): 统一 LLM JSON 调用层 llm_json_call，按域 Schema 迁移 chat/memoir agents

2026-04-03 13:44:11 +08:00

test_story_time_hints.py

feat(api+app): 对话阶段化、回忆录流水线与客户端会话体验

2026-03-26 12:13:36 +08:00

test_transcript_for_judge.py

feat(eval): internal-eval stack, judge fixes, and eval web overhaul

2026-04-07 17:18:47 +08:00

test_user_export_markdown_split.py

refactor(chat): AI-native prompts, remove interview heuristics

2026-04-06 22:23:46 +08:00

test_whisper_local.py

feat(memory,conversation): 记忆富化/证据包、时间线幂等字段与对话分段全链路

2026-03-27 16:24:43 +08:00