PagerDuty MCP Integration

Weldable's PagerDuty MCP integration connects your AI agents to PagerDuty for managing incidents, on-call schedules, and escalation policies through natural language. PagerDuty has been recognized as a Leader and Outperformer in the 2026 GigaOm Radar for IT Incident Response Platforms for the fourth consecutive year, and its AI agent suite now includes specialized agents for SRE, incident scribing, and shift management. Weldable ties PagerDuty into your broader toolchain so your agents can coordinate incident response across monitoring, communication, and project management systems.

Your agent handles requests in plain English. Tell it to trigger an incident, check who is on call, or acknowledge an alert, and Weldable maps the intent to the correct PagerDuty API endpoint.

Use cases

Automated incident creation from monitoring

When your agent detects an anomaly in Datadog or New Relic, it creates a PagerDuty incident with the alert details, affected service, and severity level pre-filled. The incident routes through the correct escalation policy based on the service, paging the right on-call engineer without anyone manually assessing and forwarding the alert. This cuts the time between detection and human response.

On-call schedule queries

Your agent checks who is currently on call for a given service or team and posts the information to Slack. This is useful for engineers who need to know who to contact about a specific system, or for managers who want a daily summary of on-call coverage across all teams. No one has to log into PagerDuty to look up the schedule.

Incident status updates

During an active incident, your agent posts regular status updates to a Slack channel by pulling the latest notes and status from PagerDuty. It tracks acknowledgment times, escalation events, and resolution progress. Stakeholders stay informed without joining the incident call or repeatedly checking PagerDuty for updates.

Post-incident reporting

After an incident resolves, your agent collects the timeline from PagerDuty: when it was triggered, who acknowledged it, what escalations occurred, and how long resolution took. It formats this into a post-incident report and posts it to Google Docs or Slack. The report includes response time metrics that feed into SLA tracking and retrospective discussions.

Escalation policy management

Your agent reads current escalation policies and identifies gaps, such as services without a secondary on-call or policies that route to inactive users. It generates a report highlighting these issues so operations teams can address coverage gaps before they cause missed alerts during real incidents.

How it works

Connect your PagerDuty account through an API key or OAuth. Weldable stores credentials securely and handles authentication for every request. Your agent gets access to incidents, services, schedules, and escalation policies based on the permissions you grant.

Describe what you need in natural language. Weldable resolves service names, escalation policy IDs, and user references automatically. Your agent can chain PagerDuty actions with other integrations: create an incident, post to Slack, and open a Jira ticket in one coordinated workflow.

Tips

Map services to your actual architecture. PagerDuty services should correspond to the systems your team owns. When your agent creates incidents, specifying the correct service ensures the right escalation policy triggers and the right person gets paged.

Urgency levels control notification behavior. PagerDuty distinguishes between high and low urgency incidents. High urgency pages immediately through phone and SMS, while low urgency sends email or push notifications. Tell your agent which urgency to use based on the severity of the issue to avoid unnecessary midnight phone calls.

Acknowledge before resolving. PagerDuty's incident lifecycle goes from triggered to acknowledged to resolved. Your agent should acknowledge incidents first to stop escalation timers, then resolve after the fix is confirmed. Skipping acknowledgment can trigger unnecessary escalations.

Incident notes provide context. Your agent can add notes to an incident with diagnostic details, links to dashboards, or steps already taken. These notes persist on the incident timeline and are valuable during post-incident reviews.

Schedule overrides handle planned absences. When a team member is out, your agent can create schedule overrides to temporarily assign their on-call shifts to a backup. This prevents pages from going to someone who is unavailable.