Operations · Advanced
On-call incident runbook investigator
Diagnose production alerts by querying metrics, reviewing recent commits, and drafting mitigations
- When something happensAlert notification received
- ThenAnalyze telemetry and inspect commits
- ResultAwaiting approval to post report
When a production alert fires, this bot executes diagnostic runbook checks against telemetry dashboards and inspects recent code deployments. It correlates error surges with commit changes, generates an incident triage summary, and proposes CLI remediation commands for engineer approval.
Prompt to start with
BotBentoRecommended
When an incident alert triggers, query Datadog for error rates, throughput spikes, and affected service shards. Review recent commits on GitHub affecting the impacted service endpoints. Compile findings into a structured incident diagnostic summary. If a remediation command needs to be executed via terminal, draft the exact command and pause for operator confirmation before running any destructive or state-changing action. Never post the incident report to Slack or execute infrastructure changes without explicit user approval.