Instant download

Nerve Ops Playbook

Use this to run agent ops, queue review, and incident triage from a repeatable operating rhythm instead of scattered memory.

Daily checks Queue triage Incident flow Reliability debt

Daily review loop

  • Check lead routing counts and stage drift.
  • Check cron freshness and suspicious output count.
  • Check queue depth by status, not just total count.
  • Check one proof-bearing artifact from a real workflow.

Incident triage rules

  • Separate “service active” from “workflow healthy.”
  • Classify failures as data, routing, execution, or publishing.
  • Patch the source path first, not the symptom.
  • Write one rollback or fallback note before a risky fix.

Queue cleanup rules

  • Archive stale work that predates the current operating model.
  • Encode repeatable filtering rules into CLI or jobs, not manual habits.
  • Separate high-fit review items from low-actionability noise.

Reliability debt review

  • List stale jobs.
  • List suspicious outputs.
  • List queue statuses with no clear owner.
  • Turn the top recurring failure into a deterministic wrapper or gate.

Upgrade path

If the playbook surfaces structural ops debt, missing modules, or cross-system friction, move into a Systems Diagnosis or a full systems build.