Overview
Go beyond simple incident lookups. This agent analyzes your PagerDuty data to answer complex questions like "What's causing the increase in incidents?" or "Which services have the most on-call fatigue?"
**Capabilities:**
- Analyze incident patterns and identify root causes
- Surface on-call workload imbalances and fatigue indicators
- Compare incident frequency across services and teams
- Identify recurring issues and proactive remediation opportunities
- Synthesize insights from incidents, services, and on-call activity
[{"type": "user", "content": "What's driving the 60% increase in PagerDuty incidents this month? We need to identify the root causes."}, {"type": "thinking", "content": "Analyzing incident patterns...", "thinkingSteps": ["Retrieving incidents from last 30 days...", "Calculating incident frequency by service...", "Identifying recurring issues and patterns...", "Comparing to historical benchmarks..."]}, {"type": "assistant", "content": "I analyzed your PagerDuty incident data and found the drivers of the increase:\n\n## Key Metrics\n\n| Metric | Value | Change |\n|--------|-------|--------|\n| Total Incidents | 142 | +60% vs last month |\n| Avg Resolution Time | 2.8 hrs | +40% vs baseline |\n| On-Call Alert Load | 18 alerts/day | +80% vs last month |\n\n## Incident Breakdown\n\n| Service | Incidents | % of Total | Primary Cause |\n|---------|-----------|------------|---------------|\n| Payment API | 58 | 41% | Database connection timeouts |\n| User Auth | 34 | 24% | Redis cache failures |\n| Notification Service | 28 | 20% | Rate limit exceeded |\n| Other | 22 | 15% | Various |\n\n## Root Causes\n\n**Primary Driver:** **Payment API** incidents increased 180% (21 → 58). 78% are **database connection timeout** errors. The database connection pool is exhausted during peak hours (avg 95% utilization vs. 60% target). This is a capacity issue, not a code bug.\n\n**Secondary Issue:** **User Auth** incidents increased 120% (15 → 34). 68% are **Redis cache failures**. The Redis cluster is hitting memory limits (92% utilization). Cache eviction is causing authentication slowdowns.\n\n**Pattern Identified:** 45% of incidents occur during **peak hours** (10 AM - 2 PM EST), correlating with traffic spikes from the new mobile app release.\n\n**Recurring Issues:** 8 incidents are **repeats** of the same database timeout issue (INC-4521, INC-4589, etc.). No permanent fix has been implemented.\n\n**Recommendation:**\n1. Scale database connection pool - increase from 50 to 100 connections, target 1 week\n2. Expand Redis cluster memory - add 2 nodes, target 3 days\n3. Implement circuit breaker for Payment API - prevent cascade failures\n\n<ENTITY_ANNOTATION id=\"INC-4521\" title=\"Payment API Database Timeout\" integrationType=\"PAGERDUTY\" />\n\n<CTA label=\"View All Incidents\" action=\"Show all incidents from this month\" />\n<CTA label=\"Generate Remediation Plan\" action=\"Create an incident reduction action plan\" />"}]
