PagerDuty MCP Server

What the PagerDuty MCP server does

How models use it and what it is built for.

The server covers PagerDuty’s core objects in sixteen tool groups, split into read and write operations: incidents, services, schedules, teams, escalation policies, event orchestrations, alert grouping, status pages, users, change events and activity logs. The read side is where most of the value is. “What changed before this incident started” joins change events to the incident timeline, and “who is on call for checkout right now” resolves a schedule through an escalation policy. The write side can manage incidents and services, so scope it deliberately. PagerDuty does not offer tool filtering on its hosted server; you restrict what an agent can do with a scoped OAuth client instead.

Tools the PagerDuty MCP server exposes

Typical tools an AI model can call. Exact names vary by version.

• incidents — list, read and manage incidents and their timelines
• services — read and manage services
• schedules, teams and escalation policies — resolve who is on call
• change events — recent deploys and changes linked to services
• event orchestrations and alert grouping — read routing and grouping setup
• status pages, users and activity logs

Connecting to PagerDuty

Taken from the official PagerDuty documentation — see PagerDuty MCP Server integration guide for the full reference.

https://mcp.pagerduty.com/mcp

https://mcp.eu.pagerduty.com/mcp

{
  "mcpServers": {
    "pagerduty": {
      "type": "http",
      "url": "https://mcp.pagerduty.com/mcp",
      "headers": {
        "Authorization": "Token token=<PAGERDUTY_API_KEY>"
      }
    }
  }
}

Example prompts to try

Copy any of these into MCP Agent Studio after connecting.

• Who is on call for the checkout service right now, and who is next in the escalation policy?

• Summarise incident #4821: timeline, responders and what changed in the hour before it fired.

• Which services had the most high-urgency incidents this month?

• List open incidents with no acknowledgement in the last 15 minutes.

Models on MCP Playground

This is not a single-model product: you get the same MCP connection with 60+ models (Claude, GPT, Gemini, DeepSeek, open-weight, and more), you can switch mid-conversation, and you can open Compare mode to run the same prompt against multiple models at once. The card above is a suggested starting point for this server — not the only choice.

Check an AI agent can actually use the PagerDuty MCP server

Listing tools proves the server is reachable, not that a model can work with it. Evals go further: they read every tool on the server, write a test suite from its real schemas, and run it — code decides pass/fail on the responses (schema conformance, error codes, pagination, result caps) while a scoring model grades plain-English tasks driven through the tools.

PagerDuty MCP server — FAQ

Common questions about connecting, scoping and using it safely.