Monitoring · 1 skill

PagerDuty MCP and Claude skills

Pull open incidents, on-call schedules and response times from PagerDuty, and post the on-call handoff or a weekly review of what paged you and why.

Connect PagerDuty to your agent (MCP)

PagerDuty MCP server, PagerDuty's official hosted server at https://mcp.pagerduty.com/mcp (https://mcp.eu.pagerduty.com/mcp for EU accounts), also in Claude's connector directory. It takes a user API key in the header Authorization: Token token=<key>, or OAuth with static client credentials (it doesn't support dynamic client registration). In Claude Code: PAGERDUTY_API_KEY=<key> claude mcp add-json pagerduty '{"type":"http","url":"https://mcp.pagerduty.com/mcp","headers":{"Authorization":"Token token=${PAGERDUTY_API_KEY}"}}'; in Cursor, add the URL under mcpServers in .cursor/mcp.json (guide). Its browse_* tools read incidents, schedules and on-calls, services, teams and logs; its manage_* tools change them. The older self-hosted server is deprecated.
REST API with a read-only key. An admin creates a general access key under Integrations > Developer Tools > API Access Keys and ticks Read-only API Key. Store it in PAGERDUTY_API_KEY and send Authorization: Token token=<key> with Accept: application/vnd.pagerduty+json;version=2.

The hosted server can't hide its write tools, so the credential decides: a read-only key allows only GET calls, and every manage_* tool fails with it.

1 PagerDuty skill

PagerDuty weekly incident report

Every Monday: last week's PagerDuty incidents by service and urgency, mean time to acknowledge and resolve, and the noisiest service.

DevOpsMondays at 9:00Report card

What agents do with PagerDuty

PagerDuty's REST API (https://api.pagerduty.com) gives agents incidents, on-calls and services. GET /incidents filters on statuses[] (triggered, acknowledged, resolved), urgencies[], service_ids[], team_ids[] and a since and until window, and each incident carries incident_number, title, status, urgency, service, assignments, created_at and last_status_change_at. GET /oncalls says who is on call for each escalation policy and level in a time range, and GET /log_entries holds every incident's timeline. The Analytics API (POST /analytics/metrics/incidents/all, with variants per service and per team) returns incident counts and mean times to acknowledge and resolve.

On-call teams schedule a morning handoff (open incidents, who is on call today and next), high-urgency incidents still unacknowledged, a weekly review with incidents by service, after-hours pages and response times, and a list of the noisiest services.

pagerduty.com

Teams that use these

Your PagerDuty numbers, on one board with everything else

Skills are free to use with any agent. Briefkin is where their results land: one board for every report, number and question your agents send, with a ping when something needs you.

Join the open beta Browse all skills

Free during the open beta. No card.