3 Commits

Author SHA1 Message Date
ymq
6707cabedc refactor(team-communication): 问题冒泡改为显式路由,规范移交 skill
规范(团队角色/问题类型/冒泡路径)移交 team-communication skill,由 LLM 读 skill 决策;
系统只固化通用问题工具,不读配置表、不存路径快照:
- communication.py: raise/resolve/escalate/list_for 显式路由,
  escalate 显式传 next_handler;current_handler_role/agentid 两列确定性过滤
- questions.py: 向后兼容层适配
- agent_loop.py: need_info/review_reject/fault_report 显式 first_handler_role
- sdlc_ability.py: escalate_question 显式转 customer
- models: 去 escalation_path/pos,加 current_handler_role/agentid
2026-08-16 22:08:00 +08:00
ymq
4f2c6da6cf feat(team-communication): 通用问题冒泡引擎(多agent感知)
- communication.py: raise/resolve/escalate/list_for,问题类型→冒泡路径,处理方=role+agentid
- questions.py: 向后兼容层,委托 communication
- agent_loop.py: 角色提问(need_info)/PM退回(review_reject)/故障(fault_report)走 raise_problem;
  _build_qna_section 按 handler 路由注入;PM approved 只答结 review_reject
- sdlc_ability.py: list_questions/answer_question/escalate_question 按 handler 路由
- models: pipeline_agent_questions 加 from_agentid/problem_type/escalation_path/escalation_pos
2026-08-16 21:25:39 +08:00
4cb406003e feat(v3.2.0): 主agent+角色agent架构 — role任务认领/问题回路
- pipeline_tasks 增加 role/claimed_by 字段+索引(存量库需ALTER)
- agent_loop 重写为角色循环:原子认领(role匹配+令牌校验)→LLM执行→交付/need_info提问
- 角色别名归一化(developer→develop等),ROLE_PROMPT改replace防花括号崩溃
- questions.py 新增问题回路:agent_ask/answer_question/forward_question/list_questions
- init.py 注册 pipeline_role_submit 与问题回路 env 接口
2026-08-05 12:58:52 +08:00