Description
SKT-0112 brought the package under the timeout, but main CI just race now reports cmd/synthkit-control-dash at 386 s (run 39082154382, 2ff0839) and a 600.302 s timeout (run 39083496929, c5ddebc, running TestLoadScenariosEnumeratesBlueprints), with 348-353 s on the gate hosts. Small CI runners plus other package load push it over. Cut the package’s own test cost (shared fixtures, fewer full-estate derivations, bounded parallelism) without skipping under race, raising the timeout or weakening any assertion.
Acceptance Criteria
- #1 Main CI just race reports cmd/synthkit-control-dash under 300 s on two consecutive main SHAs, with the unchanged 600 s timeout
- #2 No test skipped under race, no assertion removed, no timeout raised; the changed tests are named with before and after times
Definition of Done
- #1 just check (fmt-check, lint, gen-check, env-check, docs-check, test, race, hygiene, ui-check, compose-check, helm-test, lab-check, signal-fidelity)
- #2 just gen (only if a blueprint field, construct/workload config struct, or a skill under plugins/synthkit/skills/ changed)
- #3 just dump — inventory diffed against signals/
Implementation Notes
Landed 8e7f233 (loop50 BATCH11). Main CI control-dash race 189.2s on 8e7f233 and 193.7s on e8c4a8e, under 300s on two consecutive SHAs, timeout unchanged.