Skip to content
Flows
Incident Response Runbook for Agent Mishaps
Agent QA & Security60-90 minutes
0/4 steps0%

Step 1 of 4

Define detection and the kill switch

You cannot respond to what you cannot detect or stop.

Prompt capsule

Build detection and the emergency stop. DETECTION: enumerate mishap signals - blast-radius limits tripping, security events (blocked/escalated actions spiking), anomaly flags from observability (runaway cost, error bursts), audit-trail alerts on sensitive-resource changes, and human reports; define where each surfaces and who is notified. KILL SWITCH: implement a single, fast action that immediately halts the agent loop (no more turns), cancels in-flight tool executions at safe boundaries, and revokes the agent's tool authority (all tools drop to blocked/ask) - reachable by an operator without a code change and, ideally, auto-triggerable by severe signals (e.g. blast-radius hard stop). Test the kill switch: trigger it mid-task and confirm the agent stops acting within seconds and cannot execute further tools.

Paste into Claude · Complete implementation prompt with explicit requirements

Expected after this step

Detection signals wired and a fast, tested kill switch.

Should not happen

  • No kill switch, so a misbehaving agent keeps acting while you scramble
  • Improvised response that makes things worse (deleting evidence, partial rollback)
  • Investigations that cannot reconstruct what happened for lack of records
  • Postmortems that assign blame and change nothing structural

Verify before continuing

Do not move on until every check is true. The complete button stays locked until then.

Do not continue if…

  • !No kill switch, so a misbehaving agent keeps acting while you scramble
  • !Improvised response that makes things worse (deleting evidence, partial rollback)
  • !Investigations that cannot reconstruct what happened for lack of records
  • !Postmortems that assign blame and change nothing structural

If the AI messes this up

Use this when the AI fakes progress or breaks the feature. It forces a real fix.

The kill switch leaves in-flight actions running. It must cancel at the next safe boundary AND revoke authority so queued/next calls are blocked - stopping new turns is not enough if a destructive call is already dispatched.

Your notes for this step