Eight or more high-priority recurring CI drivers (including priority-98 database integrity check, priority-96 hypothesis score update, priority-94 evidence backfill) have all been stale since ~2026-04-02 — a shared stale date strongly suggesting a systemic root cause, not individual driver bugs.
**Frame: diagnose + fix + verify throughput. Not: do the work the drivers were supposed to do.**
Step 1: Determine why all drivers went stale around the same date. Query orchestra.db task_runs for failure patterns. Check for schema changes, code regressions, or connection pool issues pushed ~2026-04-02 that would cause systematic silent failures.
Step 2: Fix the root cause at the shared level — not individual driver hacks.
Step 3: Verify at least 3 of the stale drivers execute successfully post-fix.
Step 4: Write a postmortem summary in the spec Work Log.
**Critical stale drivers**: aa1c8ad8 (Database integrity, p98, 26d), 9d82cf53 (Hypothesis scores, p96, 26d), bf55dff6 (Trigger debates, p94, 26d), 33803258 (Evidence backfill, p94, 26d), 607558a9 (King of the Hill tournament, p97, 22d).
Do NOT create workaround tasks that do the work manually. Fix the engine.
Spec: docs/planning/specs/quest_senate_stale_driver_diagnosis_spec.md
Completion Notes
Auto-completed by supervisor after successful deploy to main