BMAD Incident Responder
DevOpsby AgentBox TeamPublished April 13, 2026
178Sign in to star this agent
0Sign in to follow AgentBox Team
2.8kruns
Try BMAD Incident Responder
Pre-computed resultsInput
API latency spike to 8s, error rate 15%
Description
Production incident management agent. Triages incidents by severity, guides root cause analysis with the 5-Whys method, and generates blameless postmortem reports with prevention action items.
Use in your project
$ agentbox pull bmad-incident-responder
$ agentbox bmad-incident-responder
$ agentbox mcp serve --agent bmad-incident-responder
Works with Claude Code, Cursor, VS Code, and any MCP client.
Sample I/O
Input
Production database connection pool exhausted at 14:32 UTC. API returning 503 for all users. 2,400 users affected. Pool size: 20 connections, max observed: 47 concurrent queries.
Output
## Postmortem: Database Connection Pool Exhaustion\n\n**Severity**: SEV1 | **Duration**: 23 min | **Users Affected**: 2,400\n\n### Timeline\n- 14:32 — Alert: 503 rate >5% on API gateway\n- 14:35 — On-call identified DB pool exhaustion\n- 14:38 — Increased pool size 20→50 (hot fix)\n- 14:40 — Recovery confirmed, 503 rate <0.1%\n- 14:55 — All-clear\n\n### 5-Whys\n1. Why 503? → Connection pool exhausted\n2. Why exhausted? → 47 concurrent queries vs 20 pool limit\n3. Why 47 queries? → Analytics dashboard running N+1 queries\n4. Why N+1? → Missing eager loading on agent_runs join\n5. Why not caught? → No load test for analytics endpoint\n\n### Action Items\n1. Fix N+1 query with eager loading (P0, Dev Lead)\n2. Increase pool default to 50, add PgBouncer (P1, DevOps)\n3. Add load test for analytics dashboard (P2, QA)
bmadincident-responseamspostmortem