Overview
The PagerDuty MCP server provides comprehensive tools for managing your incident lifecycle, from creation through resolution. All incident management operations can be performed through natural language commands to your AI assistant.Write operations require starting the server with the
--enable-write-tools flag. See Configuration for details.Viewing Incidents
List All Incidents
To view current incidents:list_incidents tool supports filtering by:
- Status:
triggered,acknowledged,resolved - Urgency:
high,low - Service: Filter to specific service IDs
- Team: Filter to specific team IDs
- Time range: Start and end dates
- User assignment:
assigned(your incidents) orteams(your team’s incidents)
Get Incident Details
For detailed information about a specific incident:get_incident tool returns:
- Current status and urgency
- Service and escalation policy
- Assignments and acknowledgments
- Timeline of status changes
- Related alerts
View Incident Alerts
List all alerts associated with an incident:list_alerts_from_incident tool provides alert details including event data, severity, and timestamps.
Creating Incidents
Create a new incident with thecreate_incident tool:
- title: Brief description of the incident
- service_id: The service experiencing the issue
- urgency:
highorlow(defaults to service settings) - body: Detailed description
- incident_key: Deduplication key for related events
- escalation_policy_id: Override the service’s default escalation policy
Managing the Incident Lifecycle
Acknowledge Incidents
Acknowledge one or more incidents to indicate you’re working on them:manage_incidents tool with status: acknowledged.
Resolve Incidents
Mark incidents as resolved once the issue is fixed:manage_incidents tool with status: resolved.
Reassign Incidents
Reassign an incident to another user:assignment field of the manage_incidents tool.
Change Urgency
Update incident urgency level:manage_incidents tool with urgency: high or urgency: low.
Escalate Incidents
Manually escalate an incident to the next level in the escalation policy:manage_incidents tool with the escalation_level field.
Adding Context to Incidents
Add Notes
Document findings, actions taken, or relevant information:add_note_to_incident tool creates a timestamped note visible to all responders.
View Incident Notes
Retrieve all notes for an incident:list_incident_notes tool returns all notes with author and timestamp information.
Collaborating on Incidents
Add Responders
Request additional help on an incident:add_responders tool can add:
- Users: Individual responders by user ID
- Escalation policies: Notify an entire escalation policy
Analyzing Incidents
View Related Incidents
Find incidents related to the current one:get_related_incidents tool uses PagerDuty’s machine learning to identify similar incidents based on:
- Service relationships
- Alert patterns
- Timing correlations
View Past Similar Incidents
Retrieve historical incidents with similar characteristics:get_past_incidents tool helps identify recurring issues and learn from previous resolutions.
Identify Outlier Incidents
Determine if an incident is unusual:get_outlier_incident tool provides:
- Occurrence frequency
- Rarity category
- Historical context
- Pattern matching information
Common Workflows
Triage New Incidents
- List triggered incidents:
Show all triggered incidents - Review details:
Get details for incident INC-1234 - Check alerts:
Show alerts for incident INC-1234 - Acknowledge:
Acknowledge incident INC-1234 - Add context:
Add note to incident INC-1234: "Investigating...."
Incident Response
- Acknowledge incident: Signals you’re working on it
- Check related/past incidents: Learn from similar issues
- Add responders if needed: Request help from specialists
- Document actions in notes: Keep team informed
- Resolve when fixed: Close the incident
Bulk Operations
Manage multiple incidents simultaneously:- Filter incidents by service and status
- Use
manage_incidentsto update all at once
Post-Incident Analysis
list_incidents to analyze incident trends and patterns.
Best Practices
Always acknowledge before working
Always acknowledge before working
Acknowledging incidents prevents duplicate work and lets your team know someone is investigating.
Document as you go
Document as you go
Add notes throughout your investigation. This helps with:
- Team coordination
- Knowledge sharing
- Post-incident reviews
- Future similar incidents
Use appropriate urgency levels
Use appropriate urgency levels
Set urgency correctly to ensure proper routing:
- High: Immediate impact, requires urgent response
- Low: Minimal impact, can be addressed during business hours
Leverage incident relationships
Leverage incident relationships
Check related and past incidents to:
- Identify root causes faster
- Learn from previous resolutions
- Spot recurring patterns
Resolve promptly
Resolve promptly
Mark incidents as resolved once the issue is fixed. This:
- Keeps metrics accurate
- Clears on-call queues
- Maintains clean incident history