Kevin
71fbd39e32
feat(api)!: memory single chain — async MemoryService, strict eval closure
...
Route all memory ingest/retrieve/enrichment/compaction through async MemoryService.
Remove legacy sync memory implementations (ingest/retrieve/compaction); Celery and
memoir Phase2 call asyncio.run into MemoryService-backed helpers.
Memoir Phase1 batch ingest uses MemoryService.ingest_transcripts_batch; drop chapters.
evidence_bundle_json mirror (Alembic 0015). Evaluation uses snapshot/link-only bundles;
raise EvidenceClosureMissing instead of partial/fallback lineage tiers.
Split memoir state into NarrativeCoverageState and InterviewControlState; delete the
_interview_meta_store adapter layer. Remove rolling-query and recent-fact fallback
settings from config and evidence assembly.
Update judges, docs, tests, and PlaygroundPage alignment.
Made-with: Cursor
2026-04-30 14:11:50 +08:00
Kevin
ac436b87a2
feat(api): 收敛对话与记忆流程边界,引入 LLM 网关与专用服务
...
- MemoryService 异步路径委托 MemoryIngestService / MemoryRetrievalService;富化派发经 MemoryEnrichmentScheduler
- WebSocket pipeline 经 ChatTurnService 与显式 DTO 编排单轮对话;回忆录片段入队由 MemoirIngestScheduler 封装
- 新增 LlmGateway(LlmUseCase),各 agent、任务与适配器对齐 ports
- 补充 memory 提示适配、runtime 类型、memory-retrieval 文档、ai-touchpoints 说明与扫描脚本及配套测试
Made-with: Cursor
2026-04-30 09:17:01 +08:00
Kevin
3121d1384d
WIP: memory system improvements (in progress)
...
Interview/chat prompt layers, reply planner, style profiles, memory
injection, interview meta store, and related tests. Work not finished.
Made-with: Cursor
2026-04-22 16:56:28 +08:00
Kevin
ccf7125473
fix(chat): 重复追问被拦截时再多问一次模型
...
防重复问句会把整段回复削成「这一段我记住了。」只剩一句套话时,用带纠偏说明的 system 再调一次 LLM,尽量避免用户只看到干巴巴_ack。仍只重试一次,并打日志与 meta 标记 duplicate_question_guard_llm_retry。
2026-04-10 15:35:34 +08:00
yangshilin
e1341c6d18
feat:
...
1. 建立问题库大纲,对应每个人生阶段槽位
2. 鼓励使用更生活化的交流语言共情与总结
3. 降低评审模型可能发生截断的概率
4. 成稿质量维度强化情感表达和上下文连贯性
2026-04-09 15:32:35 +08:00
Kevin
064ad2161d
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
...
- 访谈:新增 interview_state_hints,联动 orchestrator 与提示词
- 回忆录:story_pipeline_sync/state/memory/post_commit 与 Celery 任务调整
- 基建:开发用 celery broker、compose/development 脚本、依赖注入
- eval-web:移除数据集/实验/版本等页面与流式轮询,突出 Playground
- 文档与单测同步
2026-04-08 21:36:12 +08:00
Kevin
2a0c80987d
test/ 调整prompt,提高共情能力
2026-04-08 17:10:09 +08:00
Kevin
2fded6fbd9
refactor(chat): AI-native prompts, remove interview heuristics
...
- Drop interview_reply_length and utterance_substance; always run stage LLM
and memory retrieval when enabled; trim Settings fields and .env.example.
- Replace guided/opening prompts with compact fact blocks plus unified
behavior guidance; slim background_voice and persona to tone hints.
- InterviewAgent uses fixed chat_interview max_tokens/chars/segments.
Also includes stacked work: profile followup/extract path, evaluation rubric
and judge schema updates, transcript SPLIT handling in execution service,
user export markdown split tests, and golden case fixture.
2026-04-06 22:23:46 +08:00
Kevin
4cfa3843a7
chore/ 精简展示AI活动的日志
2026-04-03 13:49:24 +08:00
Kevin
bb16d3a5c9
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
...
访谈与阶段
- 新增 app/agents/stage_constants.py:集中 CHAT_STAGES、章节分类/顺序、阶段到默认 memoir 类别等,与 MemoirState 默认槽位顺序对齐;减少散落在 prompts 内的重复常量。
- 新增 app/agents/chat/prompt_context.py:以 ChatPromptContext 汇总 guided 系统提示所需字段(阶段、槽位、轮次、人设、记忆证据、回复长度模式、背景声线、职业等),统一走 get_guided_conversation_prompt。
- 大幅收敛 app/agents/chat/prompts_conversation.py;调整 prompts.py、stage_prompts.py、stage_detection.py;同步 interview_agent、profile_agent、helpers 与 state_schema,使对话侧构造提示的方式一致、可测。
回忆录流水线
- memoir/prompts.py 删除已迁至 stage_constants / 独立模板的大段常量与图片占位相关逻辑;classification / extraction / fidelity / narrative agents 与 orchest(全量历史仍可用于计数,注入模型时按轮次与字符上限截断)、image_prompt_fallback_disabled。
- dependencies 增加 get_llm_provider_fast(LRU 缓存,可与默认共用密钥与 base_url)。
任务与编排
- memoir_tasks:prepare_batches 注入 llm_fast;开启独立快档模型时打结构化日志。
- chapter_cover_tasks、story_image_tasks:与图片 prompt / JSON 工具路径或策略变更对齐(import 与行为一致)。
- story_pipeline_sync 等小处同步。
其它核心
- langchain_llm、text_normalize 随上述调用链微调。
开发者体验
- .cursor/settings.json:启用 redis-development、postman 插件。
测试
- 新增 test_image_prompt_policy:覆盖「禁止回退」等图片 prompt 策略。
- 更新 test_interview_prompts、test_interview_reply_length、test_experience_regressions、test_json_and_memory_utils,匹配新常量位置、json_utils 与对话/长度行为。
2026-04-02 12:00:00 +08:00
Kevin
69a673e6c6
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
...
Chat 访谈
- 新增 persona 系统(default / warm_listener / curious_guide)与 background_voice 语气层
- 回复长度由 compute_reply_plan 统一决策(brief / standard / expanded),融合信息密度启发式
- 输入净稿(input_normalize):编排层可选 rules/llm 归一用户口语后再喂模型与记忆检索
- 记忆证据注入:按用户话检索 memory evidence 并注入 prompt
Memoir 回忆录
- 口述归一(oral_normalize):segment 原文保留,story 管线取派生净稿作叙事输入
- segment 入队批次门闸:累计字数 + 最长等待秒数,减少零碎提交
- fidelity_check / prompts / narrative_agent 微调
- Alembic 0005:清理跨章节 story 外键
Infra
- Dockerfile 加入 ffmpeg
- pyproject.toml 新增依赖并同步 uv.lock
- .env.example / .env.production 补全新配置项
Tests
- 新增 test_background_voice、test_chat_input_normalize、test_experience_regressions
- 扩展 test_interview_prompts、test_interview_reply_length、test_story_route_oral_invariant
Made-with: Cursor
2026-03-31 23:55:26 +08:00