Repository navigation
fix: FleetReview #1371 P1s -- newly-skipped job blocks full re-run; atomic capture-router staging (t_06bf7c49) - #1422
Conversation
… verdict; atomic capture-router staging write (t_06bf7c49) - ci_overflow_integration.full_rerun_verdict: a job that started in the prior attempt but never started in the re-run now BLOCKs (newly_skipped); jobs skipped in both attempts stay exempt. - capture_router._default_write: write to <path>.<pid>.tmp, fsync, os.replace, so route_turn's existing-<turn_id>.md retry short-circuit never sees a truncated stage. Verified: both new regression tests fail on 667885f and pass with the fix; tests/test_ci_overflow_integration.py + test_mem0_capture_router_retry.py 41 passed.
FleetReviewBelow the merit threshold, review skipped (74 changed lines < 150 and 4 changed files < 6, no hot path touched; skipped by policy). FleetReview · reviewKind: skipped-by-policy |
FleetReviewReview: post-merge · head Post-merge review ( profile: light (rule: default light: lines 74<800, files 4<1000000, hunks 4<1000000, no hot path) · round 0 · members: B-assert-ctx, B-state, L6, C-assert-xhigh · families: anthropic,openai Confidence: 3/5 Findings
FleetReview provenance · models: B=gpt-6-sol, C=claude-code-opus-5-5 · cost: $0.33 · duration: 4m 02s · rounds: 1 · files examined: 4 |
Kanban t_06bf7c49. Two FleetReview P1 findings on #1371 @430c5ba9, both real-open on main 667885f.
Tests: one regression test each; both fail on main, pass here. 41 passed across tests/test_ci_overflow_integration.py and tests/plugins/memory/test_mem0_capture_router_retry.py.