Full read-write Claude connector for PagerDuty
Let Claude triage, acknowledge, and resolve incidents in PagerDuty
Stop tab-switching to handle an alert. Claude triages the incident, acknowledges and reassigns it to the right on-call, posts a status note your responders see, and runs resolve or escalate only once you approve.
PagerDuty is an incident response and on-call platform that routes alerts to the right responders and tracks incidents through to resolution. People want Claude to do more than read incident status: to triage alerts, update incidents, and manage on-call schedules from inside a conversation.
What is a PagerDuty MCP server?
A PagerDuty MCP server is a small hosted service that exposes PagerDuty's data and actions to Claude over the Model Context Protocol, so Claude can issues, code, and incidents inside your own account. We build it as a custom connector with separate read and write tools, and destructive actions stay gated behind your approval.
What can PagerDuty's current connector do today?
Today, any directory PagerDuty connection tends to be read-leaning or a thin slice: Claude can often list incidents and on-call info, but acknowledging, resolving, reassigning, adding notes, and schedule changes are usually limited or absent. A custom connector adds those write actions with scoped, human-in-the-loop controls, so Claude works inside the incident rather than beside it.
What does full read-write PagerDuty access unlock?
Full read-write PagerDuty access lets Claude triage a noisy alert, acknowledge and reassign an incident, post a status note your responders see, escalate to the next on-call, and adjust a schedule override once you approve. The incident handling that used to mean tab-switching happens where the incident already lives.
- Acknowledge an incident and reassign it to the right on-call responder
- Resolve an incident once mitigation is confirmed
- Add a status note or timeline entry visible to responders
- Trigger an escalation to the next level when an incident stalls
- Create a schedule override to cover a gap in on-call coverage
- Snooze a low-priority incident for a set window
Can Claude acknowledge and reassign incidents in PagerDuty?
Yes. A read-write PagerDuty connector lets Claude acknowledge an incident and reassign it to the right on-call responder, so it moves through its lifecycle instead of only reporting status. Reassignment and acknowledgement run as scoped write tools, separate from the gated resolve and escalate actions.
Can Claude add a status note to a PagerDuty incident?
Claude can add a status note or timeline entry that your responders see directly on the incident, and snooze a low-priority alert for a set window. These write actions flow inside the incident, while resolving once mitigation is confirmed waits for your confirmation before it runs.
Can Claude change on-call schedules in PagerDuty?
Claude can propose a schedule override to cover a gap in on-call coverage and trigger an escalation to the next level when an incident stalls, but both are gated actions applied only after you approve. The connector never has more access to schedules than your own PagerDuty account.
How a PagerDuty connector works
We build a small MCP server that authenticates to PagerDuty with OAuth and the narrowest scopes the work needs. Read tools (list incidents, check who is on call) and write tools (acknowledge, resolve, reassign, override) are separated, and outbound or destructive actions like resolve and escalate require your confirmation. Claude only ever has the access your own PagerDuty account already has.
Frequently asked questions
Can Claude acknowledge and resolve PagerDuty incidents, not just read them?
Will Claude change on-call schedules without my approval?
Does the PagerDuty connector get more access than I have?
Related connectors
Tell us what you run. We'll build the connector.
Pick your tool, choose what Claude should do, and get an instant scope and price. No call required.