Page X

@pagex list all critical incidents from the last 24 hours and show me their current status

Found 7 critical incidents in the past 24 hours: 1. Database connection timeout (INC-4521) - Triggered, assigned to DB team 2. API gateway 503 errors (INC-4518) - Acknowledged, Sarah investigating 3. Payment processor down (INC-4515) - Resolved 3 hours ago 4. Memory leak in auth service (INC-4512) - Triggered, unassigned 5. CDN cache invalidation failing (INC-4509) - Acknowledged, DevOps on call 6. Monitoring alert storm (INC-4507) - Resolved, false positive 7. Customer portal login failures (INC-4503) - Triggered, escalated to L2 This is an illustrative example showing how the MCP would surface incident data from your PagerDuty account.

This example uses the List Incidents tool with time and severity filters. Requires read access to your PagerDuty incidents. The MCP returns structured data that the AI formats into a readable summary. Actual incident details will vary based on your team's current on-call state and alert configuration.

@pagex get full details on incident INC-4521 including timeline and who's been paged

Incident INC-4521: Database connection timeout Status: Triggered Urgency: High Created: 2 hours ago (14:23 UTC) Assigned to: DB Operations team Timeline: - 14:23: Alert triggered from prod-db-monitor - 14:24: Auto-escalated to primary on-call (Mike Chen) - 14:26: Mike acknowledged - 14:31: Escalated to DB team lead (Sarah Park) - 14:35: Sarah joined incident channel This illustrative example shows the incident detail structure returned by the Get Incident by ID tool.

This example demonstrates the Get Incident by ID tool, which requires a specific incident identifier. Best used after listing incidents or when you already know the ID. Returns comprehensive incident metadata including escalation history and responder activity. Requires read permissions on incidents and users.

@pagex delete incident INC-4507 since it was confirmed as a false positive monitoring alert

Incident INC-4507 (Monitoring alert storm) has been deleted from PagerDuty. Warning: This action is permanent and removes the incident from your account history. The incident's timeline, notes, and escalation data are no longer accessible. Recommendation: For false positives, consider resolving incidents rather than deleting them to preserve audit trails and improve alert tuning analysis. This is an illustrative example of the Delete Incident tool's confirmation response.

This example shows the Delete Incident tool, which permanently removes incidents from your PagerDuty account. Use with caution — deletion is irreversible and removes audit history. Requires admin-level API permissions. Most teams should resolve incidents instead of deleting them to maintain compliance and retrospective data.

When Page X wins for closing out resolved incidents

A 6-person SRE team runs weekly incident reviews and needs to archive test incidents or duplicates created during on-call handoffs. Page X's delete and list tools let you script the cleanup: pull incidents by status, confirm IDs, then batch-delete the noise. This works best when you're closing 5-20 incidents per session—enough to justify the API key setup, not so many that you need a dedicated incident management workflow. The generic client request tool covers edge cases like updating custom fields or tagging incidents for audit trails. If your team creates fewer than 10 incidents per month, the manual PagerDuty UI is faster. For teams triaging 50+ incidents weekly, Page X turns a 20-minute admin task into a 3-minute Switchy prompt.

When Page X fits support teams triaging critical tickets

A 4-person support team uses PagerDuty to route P0 customer issues to engineering on-call. Page X's get-by-ID and list tools let you pull incident context mid-conversation—check who's assigned, what the current status is, or whether a duplicate incident already exists—without leaving Switchy. This shines when support handles 10-30 escalations per week and needs quick lookups during live customer calls. The API key auth means one shared credential for the team, no per-agent OAuth dance. The trade-off: if your support flow requires creating or updating incidents (not just reading them), you'll hit the generic client request tool for every write, which adds friction. For read-heavy escalation triage at small team scale, Page X closes the loop between support chat and incident state in one workspace.

When Page X speeds up shift change summaries

A 3-engineer on-call rotation hands off twice weekly and needs to summarize open incidents for the incoming engineer. Page X's list tool pulls all active incidents in one call, and the get-by-ID tool fetches details for anything that needs context. The outgoing engineer pastes the list into Switchy, asks for a summary grouped by severity, and shares it in Slack—no manual copy-paste from the PagerDuty dashboard. This works when your team has 5-15 open incidents per handoff and the summary takes under 5 minutes to generate. If your incident volume is higher or you need real-time alerting (not just handoff summaries), you're better off with PagerDuty's native integrations. For small on-call teams doing structured handoffs, Page X turns incident state into readable handoff notes without switching tools.