AgentEvalTool/tests
sinohqb 58c2ad0227
All checks were successful
CI / test (push) Successful in 3m10s
refactor(intelligent-eval): unify watchdog failures behind fail_eval seam
架构审查候选③(状态机归一):_TRANSITIONS 成为评估状态机的唯一真相。

三个 watchdog(planning 双闸 / executing 兜底 / 触发失败闸门)原先直接
row.status = FAILED 绕过转换表、手工重复写字段。收编到新的公开接缝
fail_eval(session, eval_id, reason, decision_type, context):表校验 →
repo CAS 条件写(status/plan_feedback/updated_at/completed_at 一条 SQL)
→ append_decision_log 留痕。

CAS 冲突(如用户在扫描间隙抢先取消)跳过并留日志,不当故障。API 路径
本已在表内,不动。新增 4 个契约测试,870 tests passed,零行为变化。
2026-08-21 03:35:05 +08:00
..
integration refactor(intelligent-eval): extract scheduler runtime + internalize decision-log dedup 2026-08-21 03:20:14 +08:00
unit refactor(intelligent-eval): unify watchdog failures behind fail_eval seam 2026-08-21 03:35:05 +08:00
__init__.py v0.2.0-dev: 文件管理 + 页面布局统一 + 6 个 bug 修复 2026-07-16 15:25:22 +08:00
conftest.py feat(intelligent-eval): terminal-state discipline watchdogs (ADR-0011) 2026-08-20 14:34:17 +08:00