yangshilin 9af2060259 fix: ...

1. 修复安卓部分机型顶部安全区遮挡回忆录标题的问题；
2. 降低封面图生成阈值和展示逻辑，独立封面图未生成时，使用正文图；
3. 去掉“嗯。”生硬回答，去掉不合理段首承接词；
4. 新增章节封面所需最少插图数的配置项

2026-04-16 20:42:54 +08:00

..

__init__.py

refactor(api,expo): 多智能体与会话收敛、回忆录兼容层移除、后端测试集大幅删减

2026-03-22 18:10:28 +08:00

agent_turn.py

refactor(eval+memoir)：精简内部评测路由与服务，composite/对话摘要与 judge 能力补强

2026-04-08 21:36:12 +08:00

background_voice.py

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

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

helpers.py

chore/ 精简展示AI活动的日志

2026-04-03 13:49:24 +08:00

interview_agent.py

fix(chat): 重复追问被拦截时再多问一次模型

2026-04-10 15:35:34 +08:00

interview_state_hints.py

fix(chat): 重复追问被拦截时再多问一次模型

2026-04-10 15:35:34 +08:00

interview_turn_plan.py

fix:

2026-04-10 20:35:57 +08:00

occupation_context.py

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

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

orchestrator.py

feat(chat): server-side interview turn plan (mode, anchor slot, snippet)

2026-04-10 14:34:38 +08:00

output_rules.py

fix:

2026-04-16 20:42:54 +08:00

personas.py

fix: 优化模型提示词，照顾用户情绪

2026-04-09 18:18:28 +08:00

profile_agent.py

feat: 回忆录证据血缘与内部评测可追溯，顺带对齐本地评测台与 CI

2026-04-08 15:37:09 +08:00

prompt_context.py

feat(chat): server-side interview turn plan (mode, anchor slot, snippet)

2026-04-10 14:34:38 +08:00

prompts_conversation.py

fix:

2026-04-16 20:42:54 +08:00

prompts_profile.py

refactor(eval+memoir)：精简内部评测路由与服务，composite/对话摘要与 judge 能力补强

2026-04-08 21:36:12 +08:00

prompts.py

feat: 回忆录证据血缘与内部评测可追溯，顺带对齐本地评测台与 CI

2026-04-08 15:37:09 +08:00

reply_limits.py

fix:

2026-04-16 20:42:54 +08:00

schemas.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

slot_question_bank.py

feat: 丰富问题模板库大纲

2026-04-09 15:44:14 +08:00

stage_detection.py

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

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

stage_prompts.py

test/ 调整prompt，提高共情能力

2026-04-08 17:10:09 +08:00