Workers and automated maintenance

What it covers

• Runtime loop/local agent update local state/heartbeat; apply orchestrator/queue execute approved operations; the recovery coordinator handles incomplete operations; the event journal retains events.
• Hardware worker/scheduler collect catalogue/evidence; permission baseline/integrity worker detects drift; supervisor/watchdog monitor services/workers.
• A worker finding must include generation time, source, evidence, severity, subject and a recommended plan.

How to use it

1. Open the finding.
2. Check freshness/evidence.
3. Check dependency/load.
4. Review the recommendation.
5. Approve/reject the controlled action.

Important rules

• A worker does not execute arbitrary shell.
• It does not mask failure as success.
• It does not delete evidence/audit.
• It does not block boot indefinitely.
• It does not auto-accept high risk.