docs: capture sentinel quick verify post-task lessons

2026-06-28 05:41:29 +00:00
parent 80467d7303
commit 0840bf392c
2 changed files with 4 additions and 0 deletions
@@ -123,6 +123,8 @@ Source mirror readiness must be proven by the internal mirror object/read probe

 Dashboard aggregate counters may include historical runs. When they disagree with the latest selected run, closeout should name the latest `runId` and report hash as the acceptance source, and track the aggregate-labeling UI improvement separately instead of treating historical aggregate red counts as the latest run's blocker.

+For Code Agent multi-round quick-verify, accept the latest run's `turn-summary` / `trace-frame` plus `blockingFindingCount=0` and `controlFindingCount=0`. Analyzer red findings about hydration, API-to-DOM lag or timing drift are investigation evidence unless they coincide with missing durable turns/final responses, failed submit/login/auth, broken continuity, absent report or unavailable user path.
+
 ## Multi-Sentinel Target Architecture

 Current implementation has independent sentinel runner services, but each runner still serves its own dashboard. The target architecture should split: