Developer Tools & Devops · Pagerduty + Claude
PagerDuty MCP with Claude: Faster Incident Response and Smarter On-Call Handoffs
The short answer
Claude connects to PagerDuty through the Model Context Protocol (MCP), a secure standard that lets Claude read and act inside the tool with permissions you set. Once connected, your IT team can ask for incident status, fetch on-call details, trigger alerts or update schedules in natural language, while people approve consequential actions.
What is Pagerduty?
PagerDuty is a real-time incident management platform used by IT operations, site reliability and DevOps teams. It sends alerts when systems fail, tracks who is on call, and coordinates the response until a problem is resolved. Organisations use it to reduce downtime, keep customers informed and make sure the right engineer is contacted at the right time, day or night.
How Claude connects to Pagerduty through MCP
The Model Context Protocol (MCP) is an open standard that lets Claude securely use other software. Think of it as a universal adapter between Claude and a tool like PagerDuty. PagerDuty can be connected through its own MCP server or through a secure direct API connection, using the sign-in method your organisation chooses.
Once connected, Claude can typically read open incidents, look up who is on call, summarise recent alerts and fetch service and schedule details. With permission, it can also help trigger an alert, acknowledge an incident or update a schedule. Engineers ask for these things in everyday language rather than navigating screens.
Permissions and approvals keep people in charge. Access is granted with the least privilege needed for each job. Consequential actions, such as paging a team or changing a rota, can require a named person to confirm first. Every action is logged, so you can review what was done, when and by whom.
What you can automate
Incident status summaries for leaders
A head of operations asks Claude for a plain-English summary of all open high-priority incidents, their age and who is working on each. Claude gathers the data from PagerDuty and presents it as a short briefing, saving the leader from logging in and reading dashboards during a busy outage.
Quick on-call lookups
When a customer-facing system shows errors, a support lead asks who is currently on call for that service. Claude retrieves the current on-call engineer and their escalation contact, so the right person can be reached within minutes rather than after searching through spreadsheets and chat threads.
Alert triage support
An SRE team receives a burst of alerts overnight. Claude groups related alerts, highlights the likely root service and drafts a suggested first response for the engineer to review. The engineer keeps control of every decision and any action taken.
Schedule updates for leave and cover
A team manager needs to swap a shift because an engineer is unwell. The manager asks Claude to propose a cover arrangement from the existing rota. After the manager approves, Claude prepares the schedule change for confirmation, reducing manual edits and missed handovers.
Post-incident follow-up
After an incident closes, a DevOps lead asks Claude to list the timeline of alerts, acknowledgements and escalations. Claude compiles the facts into a draft review that the team edits and approves, making post-incident reviews faster and more complete.
An end-to-end workflow: Overnight outage to customer update
TriggerA critical alert fires in PagerDuty for the payments service at 02:15.
Claude fetches the open incident and the current on-call engineer from PagerDuty and shares them with the duty manager.
Claude checks the related support queue in the ticketing system and counts customer reports linked to the same service.
Claude drafts a short incident update and a status page message, which the on-call engineer reviews.
After the engineer approves, Claude posts the internal update to the team's Microsoft Teams channel.
When the incident is acknowledged and resolved, Claude compiles the alert timeline into a draft review for the morning team.
OutcomeTypically, the duty manager has a clear picture within minutes, customers receive timely updates, and the post-incident review starts with an accurate timeline instead of reconstructed memory.
How we set it up
Free automation audit
We review how your IT and operations teams currently handle incidents, on-call handovers and alert noise, and identify the workflows worth automating first. The audit is free and carries no obligation.
Secure connection
We connect Claude to your PagerDuty account using the official MCP server or a managed connector, with sign-in through OAuth sign-in or an API key, depending on what your security team approves. Access is limited to the permissions each workflow needs.
Workflow design
We design the end-to-end workflow across PagerDuty and your other systems, such as Microsoft 365, ticketing or your CRM. We agree in advance which actions need human approval.
Testing against real cases
We test the workflow against past incidents and on-call changes from your own records. We refine prompts, permissions and approval steps until the results match what your team expects.
Launch, monitoring and training
We launch with monitoring and audit logs in place and train your engineers, managers and support staff on how to use Claude safely with PagerDuty. We stay available to adjust the workflow as your team learns.
Who benefits
- Site reliability or DevOps engineer
- Gets faster access to incident details and alert context, and spends less time switching between tools during an outage.
- IT operations manager
- Gains clearer visibility of open incidents and on-call coverage, with an audit trail for every action taken.
- Customer support lead
- Can quickly confirm who owns an issue and pass accurate updates to customers.
- Engineering or IT director
- Receives concise incident briefings and better-prepared post-incident reviews without extra admin work for the team.
Security and control
- Least-privilege access, so Claude can only use the PagerDuty permissions each workflow requires.
- Human approval for consequential actions such as paging teams, changing schedules or closing incidents.
- Full audit logs of what Claude read and did, reviewable by your security and operations leads.
- Testing against real past cases before launch, with clear escalation routes if the automation fails.
- Regular reviews of permissions and sign-in credentials, including rotation of API keys where used.
For teams in the US, UK, India and the Middle East
Organisations in the United States, the United Kingdom, India and the Middle East have different data protection obligations, including the Digital Personal Data Protection Act, 2023 in India and data protection laws in the UAE and Saudi Arabia. We design each deployment around these requirements. Claude can be hosted in India, the United States, the United Kingdom or the European Union, and the hosting region is chosen to meet your legal and contractual needs.
Learn more about Claude and MCP
Anthropic's AI model and assistant.
Secure links between Claude and business tools.
How engineers build and change software with Claude.
Hosted agents that run multi-step work.
The open standard Claude uses to connect to tools.
Frequently asked questions
What does the PagerDuty MCP server do?
The PagerDuty MCP server is a secure connector that lets Claude use PagerDuty through the Model Context Protocol. It allows Claude to read incident and on-call information and, with permission, carry out actions such as triggering alerts or updating schedules. It acts as a controlled gateway, so access is defined by the permissions you grant rather than by open access to your account.
Is connecting PagerDuty to Claude secure?
Security depends on how the connection is set up. We use sign-in through OAuth or an API key, limit access to the permissions each workflow needs, require human approval for consequential actions and keep audit logs of activity. Credentials are stored and rotated according to your security policy. We do not guarantee compliance outcomes, but we design the setup to fit your existing controls.
What can Claude do with PagerDuty once connected?
Typically, Claude can look up open incidents, identify the current on-call engineer, summarise alert history and fetch service and schedule details. With permission, it can help trigger an alert, acknowledge an incident or prepare a schedule change. The exact actions depend on the permissions you approve for each workflow.
How long does setup take?
Timelines vary with the number of services, schedules and other systems involved. A focused workflow often takes a few weeks from the free audit to launch, including testing and training. Broader rollouts across several teams usually take longer. We confirm a realistic timeline during the audit.
Which PagerDuty plan or permissions do I need?
Access depends on your PagerDuty account setup and the permissions your administrators grant to the sign-in method used. In general, the connection needs rights for the actions your workflow performs, so we recommend a dedicated integration user with least-privilege access. We confirm the exact requirements with your PagerDuty administrator during the audit.