Harness-engineering harness-incident-response

name: harness-incident-response version: "1.0.0" description: Runbook generation, postmortem analysis, and SLO/SLA tracking stability: static cognitive_mode: diagnostic-investigator triggers:

• manual
• on_bug_fix platforms:
• claude-code
• gemini-cli
• cursor
• codex tools:
• Bash
• Read
• Write
• Edit
• Glob
• Grep
• emit_interaction cli: command: harness skill run harness-incident-response args:
  • name: path description: Project root path required: false
  • name: mode description: "Operating mode: runbook, postmortem, or slo. Auto-detected when omitted." required: false
  • name: severity description: "Incident severity: sev1, sev2, sev3, sev4. Defaults to sev3." required: false mcp: tool: run_skill input: skill: harness-incident-response path: string type: rigid tier: 3 internal: false keywords:
• incident
• runbook
• postmortem
• SLO
• SLA
• SLI
• on-call
• alert
• severity
• MTTR
• MTTD
• error budget
• incident management
• PagerDuty
• OpsGenie stack_signals:
• "runbooks/"
• "docs/runbooks/"
• "docs/incidents/"
• "docs/postmortems/"
• "slo.yaml"
• ".pagerduty*" phases:
• name: assess description: Determine incident scope, severity, and affected services from available evidence required: true
• name: investigate description: Trace root cause through logs, metrics, dependencies, and recent changes required: true
• name: document description: Generate runbooks, postmortem reports, and timeline reconstructions required: true
• name: improve description: Propose SLO adjustments, alerting improvements, and preventive action items required: true state: persistent: false files: [] depends_on: []