What problem does it solve? When a production incident occurs, teams without a prepared runbook waste critical minutes deciding what to check and which commands to run. This Skill generates a structured incident response runbook from existing design and SLO documents so on-call engineers can respond quickly and consistently. ## Core Features & Use Cases - Severity Definitions: Standardizes SEV1–SEV4 classifications with numeric first-response and resolution targets. - Scenario-Based Procedures: Produces copy-paste-ready diagnostic and mitigation commands for high latency, error spikes, database failures, resource exhaustion, external API outages, and security incidents. - Rollback & Postmortem Sections: Includes application, database migration, and configuration rollback steps plus a postmortem template for root cause analysis. - Use Case: After finishing a spec-driven design for a service, run the Skill with the spec slug to produce a runbook.md containing on-call contacts, emergency kubectl/psql commands, recovery checklists, and incident closure criteria. ## Quick Start Generate an incident response runbook for the google-ad-report spec using its existing design.md and slo.md files.
