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

数据库与模型:新增多版迁移(章节证据快照、对话血缘、记忆事实/时间线 lineage 等),把「成稿 ↔ 对话/记忆」的溯源信息落到表结构里。
业务链路:会话与 WS、回忆录/故事流水线、记忆写入与 enrichment 等跟着接上线索与快照;新增章节证据快照与评测侧 EvalTraceService 等模块,方便组评审用的证据包。
内部评测:自动化 run 与手工 memoir 评审共用可追溯证据;rubric/ judge 相关脚本与文档有配套调整。
app-eval-web:Memoir/实验详情里能展开看证据摘要与 evidence_trace(含对话轮次 id);Vite 代理与 development.sh 注入的 API 端口与当前默认内部评测端口一致,避免改端口后页面连错服务。
工程杂项:GitHub Actions / 仓库说明有更新;各适配器与支付/配额/plan 等多处为小改动或跟随主改动的收尾;新增/扩充了?
This commit is contained in:
Kevin
2026-04-08 15:37:09 +08:00
parent 6772e1269c
commit 309a051038
109 changed files with 4125 additions and 858 deletions

View File

@@ -12,17 +12,17 @@ from app.agents.chat.agent_turn import AgentChatTurn
from app.agents.chat.helpers import format_history_string, get_history_with_window
from app.agents.chat.personas import normalize_interview_persona
from app.agents.chat.prompt_context import ChatPromptContext
from app.agents.chat.stage_detection import keyword_fallback_primary_stage
from app.agents.chat.prompts_conversation import (
SLOT_NAME_MAP,
get_opening_prompt,
)
from app.agents.state_schema import MemoirStateSchema
from app.agents.chat.reply_limits import (
nonempty_segments_or_fallback,
segments_from_llm_response,
truncate_chat_segments,
)
from app.agents.chat.stage_detection import keyword_fallback_primary_stage
from app.agents.state_schema import MemoirStateSchema
from app.core.agent_logging import (
agent_span,
log_agent_payload,
@@ -92,6 +92,8 @@ class InterviewAgent:
background_voice: str = "default",
normalized_user_message: Optional[str] = None,
occupation: str = "",
profile_birth_year: int | None = None,
profile_era_place: str = "",
) -> AgentChatTurn:
"""生成状态感知的访谈回复,不持久化(由 Orchestrator 负责)"""
if not self.llm:
@@ -136,6 +138,8 @@ class InterviewAgent:
memory_evidence_text=memory_evidence_text,
background_voice=background_voice,
occupation=occupation,
profile_birth_year=profile_birth_year,
profile_era_place=profile_era_place,
)
system_prompt = ctx.guided_system_prompt()
messages: List[Any] = [SystemMessage(content=system_prompt)]