developer-toolsoauth2

PagerDuty

On-call scheduling and incident response.

Verdict

PagerDuty via MCP plugs incident data, schedules, and escalation policies into the model. Useful for on-call retros, schedule audits, and incident summaries where you want to query "what happened" instead of click-hunting through PD's UI. What we notice: the model handles incident timeline reconstruction well — given an incident ID, it can pull the events, acknowledgements, and resolution notes and produce a clean timeline. Schedule queries are also a quiet win ("who's on-call for the platform team next week"). Write operations (creating incidents, modifying schedules) work but rarely need AI in the loop. Best for: post-incident review prep where the model assembles the timeline before the human meeting; on-call schedule audits ("are we covered for the next two weekends"); weekly incident summaries for leadership; correlating PagerDuty incidents with deploys or commits. Avoid for: incident response itself (the PD app and existing alerting paths are faster); workflows where MTTR is measured in seconds (LLM round-trips break that budget); compliance-bound orgs where PagerDuty audit trails matter more than convenience. Practical frame: API access included with paid PagerDuty tiers. Token cost is small — incident objects are compact. The win is "I don't have to write the post-incident timeline from scratch" — most of it is already in PD, the model just needs to read it.

Common use cases

  • Check open incidents during standup
  • Update escalation policy from chat
  • Audit recent changes to on-call schedules
  • Associate service dependencies before deploy
  • Pull incident metrics for postmortem prep

Integration

Vendor
PagerDuty
Category
developer-tools
Auth
OAUTH2
Tools
50
Composio slug
pagerduty

Tools

  • Add service to incident workflow trigger

    This endpoint adds a service to an existing incident workflow trigger in pagerduty. it allows you to associate a specific service with a trigger, enabling automated incident management processes for that service. use this endpoint when you

  • Aggregate escalation policy incident metrics

    This endpoint analyzes and aggregates incident metrics across all escalation policies in pagerduty. it allows for detailed filtering and customization of the analysis, enabling users to gain insights into incident patterns, response times,

  • Associate service dependencies

    Associates multiple service dependencies in pagerduty, allowing you to define relationships between supporting and dependent services. this endpoint is used to establish a hierarchical structure of services, which is crucial for effective i

  • Associate team with automation action

    This endpoint associates a specific team with an automation action in pagerduty. it allows you to link a team to an automated workflow, enabling better organization and management of automation actions within your incident response processe

  • Audit escalation policy records

    Retrieves the audit records for a specific escalation policy in pagerduty. this endpoint allows users to access a detailed history of changes made to the escalation policy, including modifications to escalation rules, associated services, a

  • Convert service rule by id

    Converts the rules associated with a specific pagerduty service to a new or optimized format. this endpoint is used to update and standardize the incident management rules for a given service, improving the efficiency and consistency of ale

  • Count paused incident reports

    Retrieves the count of paused incident reports in the pagerduty system. this endpoint is part of pagerduty's incident management and analytics features, providing a quick way to quantify the number of incidents that are currently in a pause

  • Create a custom field for an incident type

    Create a custom field for an incident type. custom fields (cf) are a feature which will allow customers to extend incidents with their own custom data, to provide additional context and support features such as customized filtering, search

  • Create a field option for a custom field

    Create a field option for a custom field. custom fields (cf) are a feature which will allow customers to extend incidents with their own custom data, to provide additional context and support features such as customized filtering, search an

  • Create and manage custom incident fields

    Creates a new custom field for incidents in pagerduty. this endpoint allows you to define additional data points to be associated with incidents, enhancing incident management and reporting capabilities. custom fields can be used to store v

  • Create an incident type

    Create a new incident type. incident types are a feature which will allow customers to categorize incidents, such as a security incident, a major incident, or a fraud incident. <!-- theme: warning --> > ### early access > this endpoint is i

  • Create a ruleset

    Creates a new ruleset in pagerduty for managing incident routing and notification rules. this endpoint allows you to define a named set of rules that determine how incidents are processed and directed to specific teams or users. it's partic

  • Create a trigger

    Create new incident workflow trigger scoped oauth requires: `incident workflows.write`

  • Create automation runner endpoint

    Creates a new automation runner in the pagerduty incident management system. this endpoint allows you to set up either a sidecar runner, which polls for invocations externally, or a runbook runner, which communicates directly with a runbook

  • Create business service

    This endpoint creates a new business service in pagerduty, which represents a specific service or application that can be monitored and managed within the incident management platform. it allows you to define essential properties of the ser

  • Create escalation policy

    Creates a new escalation policy in pagerduty, defining how incidents are escalated to different responders or teams. this endpoint allows you to set up a structured response plan for managing incidents, including escalation rules, targets,

  • Create event orchestration

    Creates a new event orchestration in pagerduty, which defines how incoming events are processed and routed to appropriate services. this endpoint allows you to set up a configuration that manages the flow of events from various integrations

  • Create event rule in ruleset

    Creates a new event rule within a specified ruleset in pagerduty's incident management system. this endpoint allows you to define complex conditions for event matching and specify actions to be taken when an event meets those conditions. us

  • Create extension object

    Creates a new extension in pagerduty, allowing for additional functionality or integrations to be added to your services. this endpoint is used to set up webhooks, custom integrations, or other specific features that enhance the capabilitie

  • Create handoff notification rule

    Creates a new on-call handoff notification rule for a specific user in pagerduty. this endpoint allows you to set up automated notifications for when a user is about to start or end their on-call shift. it defines how and when the user shou

  • Create incident field option

    Creates a new field option for a specific custom field in pagerduty's incident management system. this endpoint allows you to add predefined options to custom fields, enhancing the ability to categorize and filter incidents with consistent,

  • Create incident record

    Creates a new incident in pagerduty with specified details and assignments. use this endpoint when an event requires immediate attention from on-call teams. it supports comprehensive incident documentation, including priority, urgency, and

  • Create incident workflow

    This endpoint creates a new incident workflow in pagerduty, allowing users to define a series of automated steps to be executed during incident response. it enables the setup of complex, multi-step processes that can include actions like se

  • Create incident workflow instance

    Creates a new instance of an incident workflow for a specific incident in pagerduty. this endpoint allows you to initiate a predefined workflow process for managing and resolving an ongoing incident. it links the workflow instance to a part

  • Create integration for orchestration

    Creates a new integration within an existing event orchestration in pagerduty. this endpoint allows users to set up connections between pagerduty and external systems or services, enabling automated event management and incident response. i

  • Create integration for service

    This endpoint creates a new integration for a specific pagerduty service. it allows you to set up various types of integrations, including email-based and events api v2 integrations. the primary use case is to establish new channels for inc

  • Create maintenance window object

    Creates a new maintenance window in pagerduty, allowing you to schedule periods of planned maintenance for specific services. during a maintenance window, incidents are not created for the affected services, preventing unnecessary alerts an

  • Create new schedule layer

    Creates a new on-call schedule in pagerduty, defining rotation patterns for team members. this endpoint allows you to set up complex schedules with multiple layers, each having its own rotation pattern, user assignments, and time restrictio

  • Create new service

    Creates a new service in pagerduty, which is a key component for managing incidents and alerts. this endpoint allows you to set up a service with various configurations including escalation policies, incident urgency rules, support hours, a

  • Create new tag in system

    Creates a new tag in pagerduty for labeling and categorizing entities. this endpoint allows users to add a custom label to their pagerduty account, which can be used to organize and filter various resources such as incidents, services, or t

  • Create new team with details

    Creates a new team within your pagerduty account. this endpoint allows you to establish a group of users with shared responsibilities, often used for organizing incident management and on-call schedules. use this when setting up a new depar

  • Create new webhook subscription

    Creates a new webhook subscription in pagerduty to receive specific event notifications. this endpoint allows users to set up automated notifications for various incident-related events, such as triggers, acknowledgments, and resolutions. t

  • Create or update status page post

    Creates a new post on a specified pagerduty status page. this endpoint allows you to publish information about incidents or scheduled maintenance, including details such as the post type, timing, and associated updates. use this to communic

  • Create or update status page postmortem

    Creates or updates a postmortem for a specific status page post in pagerduty. this endpoint allows you to add detailed analysis and insights about an incident after its resolution. the postmortem can include rich-text content and offers the

  • Create response play for incidents

    Creates a new response play in pagerduty, which is a predefined set of actions to be executed in response to an incident. this endpoint allows you to configure automated responses, including adding subscribers, assigning responders, setting

  • Create schedule overrides configuration

    Creates one or more overrides for a specific schedule in pagerduty. this endpoint allows you to temporarily modify the on-call schedule by assigning different users for specific time periods. it's useful for handling planned absences, shift

  • Create service event rule

    Creates a new event rule for a specific service in pagerduty. this endpoint allows you to define complex conditions and actions for handling incoming events, enabling automated incident management and alert routing. use this when you need t

  • Create statuspage subscription

    Creates a new subscription for a pagerduty status page. this endpoint allows users to set up notifications for specific status pages or components within those pages. it's used to keep subscribers informed about updates, incidents, or chang

  • Create status update template

    Creates a new template in pagerduty for standardized status updates. this endpoint allows users to define customizable templates for various notification types, including emails and short messages. it's primarily used to streamline communic

  • Create user notification rule

    Creates a new notification rule for a specific user in pagerduty. this endpoint allows you to define how and when a user should be notified about assigned incidents. you can specify the delay before notification, the method of contact, and

  • Create user object

    Creates a new user in the pagerduty system with the specified attributes. this endpoint should be used when onboarding new team members or adding users to your pagerduty account. it allows you to set up essential user information, including

  • Create user status update notification rule

    Creates a new status update notification rule for a specific user in pagerduty. this endpoint allows you to define how a user will be notified about incident status updates. it's used to customize notification preferences for individual use

  • Create workflow integration connection

    Create a new workflow integration connection. scoped oauth requires: `workflow integrations:connections.write`

  • Delete account subscription by id
    destructive

    Deletes the account subscription associated with a specific business service in pagerduty. this endpoint is used to terminate the subscription or billing relationship between a business service and the account. it should be used when you ne

  • Delete a custom field for an incident type
    destructive

    Delete a custom field for an incident type. custom fields (cf) are a feature which will allow customers to extend incidents with their own custom data, to provide additional context and support features such as customized filtering, search

  • Delete addon by id
    destructive

    Deletes a specific add-on from the pagerduty account using its unique identifier. this endpoint should be used when you want to remove an integration or extension that is no longer needed or active in your pagerduty setup. it's important to

  • Delete a field option for a custom field
    destructive

    Delete a field option for a custom field. <!-- theme: warning --> > ### early access > this endpoint is in early access and may change at any time. you must pass in the x-early-access header to access it. scoped oauth requires: `custom fiel

  • Delete alert grouping setting
    destructive

    This endpoint allows you to delete specific alert grouping settings in pagerduty. it is used to remove outdated or unnecessary configurations that determine how alerts are grouped into incidents. the deletion is permanent and cannot be undo

  • Delete all oauth delegations
    destructive

    Delete all oauth delegations as per provided query parameters. an oauth delegation represents an instance of a user or account's authorization to an app (via oauth) to access their pagerduty account. common apps include the pagerduty mobile

  • Delete automation action by id
    destructive

    Deletes a specific automation action from the pagerduty system. this endpoint is used to permanently remove an automation action, which includes any associated scripts or jobs in runbook automation. it should be used when an automation acti

Setup

Setup guide

  1. 11. In Switchy, open your workspace settings and navigate to the MCP integrations panel. 2. Click 'Add Integration' and select PagerDuty from the developer tools category. 3. You'll be redirected to PagerDuty's OAuth consent screen — sign in with an account that has admin permissions on your PagerDuty instance. 4. Grant the requested scopes: read/write access to incidents, services, escalation policies, schedules, teams, and analytics data. 5. After authorizing, Switchy redirects you back and confirms the connection is live. 6. Open any Space and type '@PagerDuty list open incidents' to test — the AI should return current incident data. 7. If the response is empty or errors, verify your PagerDuty account has active services and that the OAuth token hasn't expired in PagerDuty's app settings.

What teammates see: by default, memories from PagerDuty are scoped to the Space (PROJECT visibility) - you can mark any memory PRIVATE or share it ORG-wide.

Works well with

Top models

Compatibility data appears once enough Spaces have used this MCP together with a given model.

How Switchy teams use it

Not enough Spaces yet to publish anonymised usage stats (we require ≥ 50 Spaces per week).

Starter prompts

List Open Incidents

@PagerDuty show me all open incidents right now, including their urgency level and who's assigned to each one.
Open in a Space →

Audit Escalation Policy

@PagerDuty get the audit log for the 'Backend Services' escalation policy, focusing on changes made in the last 30 days.
Open in a Space →

Add Service Dependency

@PagerDuty associate the 'Payment API' service as dependent on the 'Database Cluster' service so incidents cascade correctly.
Open in a Space →

Count Paused Reports

@PagerDuty how many incident reports are currently paused? I need to know if we have a backlog of unresolved postmortems.
Open in a Space →

Aggregate Policy Metrics

@PagerDuty aggregate incident metrics for all escalation policies over the last quarter, breaking down by response time and resolution rate.
Open in a Space →

Example outputs

Illustrative - representative of the model's voice and quality, not literal recordings.

Prompt

@PagerDuty show me all critical incidents from the last 48 hours and summarize which services are having the most issues

Output

Found 12 critical incidents in the past 48 hours. The API Gateway service accounts for 5 incidents (42%), primarily timeout errors between 2-4 AM UTC. Database Cluster had 4 incidents related to connection pool exhaustion. Payment Processing service had 3 incidents, all during peak traffic windows. Average time-to-acknowledge was 8 minutes across all incidents. Two incidents remain unresolved: INC-9847 (API Gateway timeout) and INC-9851 (Database connection issue).

Notes

This example demonstrates read-only incident querying and AI synthesis of patterns. Requires incidents:read scope. The MCP returns structured incident data; the AI identifies trends across services. Note that incident counts and patterns reflect your actual PagerDuty account state at query time—results will vary based on your organization's recent activity.

Prompt

@PagerDuty create a new incident for the checkout service with high urgency, title 'Payment gateway returning 503 errors', and assign it to the payments on-call team

Output

Created incident INC-9852 with high urgency. Title: 'Payment gateway returning 503 errors'. Assigned to: Payments On-Call (current responder: Sarah Chen). Incident URL: https://yourorg.pagerduty.com/incidents/INC-9852. Notification sent via SMS and push to on-call engineer. Escalation policy will trigger in 15 minutes if not acknowledged. The incident is now visible in your PagerDuty dashboard and mobile app.

Notes

This write action creates a real incident in your PagerDuty account and triggers notifications to on-call responders. Requires incidents:write scope. Use cautiously—this will page actual team members. Consider testing in a sandbox service first. The MCP confirms creation but doesn't prevent duplicate incidents if you run the same prompt twice.

Prompt

@PagerDuty pull the escalation policy for our database team and draft a postmortem section explaining how alerts would have routed during last night's outage

Output

The Database Team escalation policy has three levels: Level 1 escalates to the primary on-call (5-minute timeout), Level 2 adds the secondary on-call and database lead (10-minute timeout), Level 3 pages the entire infrastructure team and CTO. During last night's outage (22:15-23:40 UTC), the initial alert would have paged the primary on-call at 22:15. With no acknowledgment, the secondary and database lead would have been paged at 22:20. The full infrastructure escalation would have triggered at 22:30—10 minutes before the issue was actually resolved. This suggests the escalation timing was appropriate, though Level 2 response could be tightened to 7 minutes to reduce time-to-engagement.

Notes

This example combines escalation policy retrieval with AI reasoning about incident response timing. Requires escalation_policies:read scope. The MCP fetches policy structure; the AI applies it to a hypothetical scenario. Useful for postmortems and policy reviews, but the AI's timing analysis is illustrative—actual paging logs would provide definitive data on who was notified when.

Use-case deep-dives

Post-incident review at 8-person SRE team

When PagerDuty MCP closes the loop on incident retrospectives

A platform team running 40 microservices wants to pull incident timelines, escalation paths, and service dependency maps into their weekly retro doc without tab-switching. The PagerDuty MCP wins here because it surfaces audit trails and escalation policy metrics directly in the AI workspace—your team can ask "which services escalated most last sprint" and get the answer inline. The OAuth2 flow means each engineer authenticates once; after that, queries run against live PagerDuty data without manual exports. The threshold: if your team only reviews incidents once a quarter, the setup overhead outweighs the benefit. But for weekly or bi-weekly retros where you're correlating incidents with deploys or customer complaints, this MCP turns PagerDuty's 50-tool API surface into a conversational interface that actually gets used.

On-call handoff documentation for distributed team

Why this MCP matters for async on-call transitions

A 12-person engineering org spread across three time zones hands off on-call every week. The outgoing engineer needs to brief the incoming one on paused incidents, active escalation policies, and any workflow triggers that fired overnight. The PagerDuty MCP lets you generate a handoff summary by asking "what incidents are paused and which services had escalations in the last 48 hours"—no need to screenshot dashboards or copy-paste from Slack threads. The 50 available tools cover incident state, service dependencies, and automation actions, so the summary includes context the next engineer actually needs. The catch: if your on-call rotation is stable and incidents are rare, a static runbook is simpler. But for teams where on-call context changes daily and handoffs happen across time zones, this MCP turns PagerDuty's data into a living handoff doc that updates itself.

Customer support escalation triage at 20-seat company

When to skip this MCP for support-to-engineering handoffs

A support team at a B2B SaaS company wants to escalate customer-reported outages to engineering without guessing which service or team to page. The PagerDuty MCP can query service dependencies and escalation policies to route the ticket correctly, but here's the trade-off: if your support team isn't already fluent in PagerDuty's service taxonomy, the MCP won't magically teach them which service maps to which customer feature. The 50 tools assume you know what you're looking for—"associate service dependencies" or "aggregate escalation policy metrics" are useful if you're already managing incidents in PagerDuty, less so if you're just trying to figure out who to ping. For teams where support and engineering share a common incident language and PagerDuty is the source of truth, this MCP saves 10 minutes per escalation. For teams still building that shared vocabulary, start with a simpler integration.

Frequently asked

What does the PagerDuty MCP let me do in Switchy?

It connects your PagerDuty incident management workflows to Switchy's AI workspace. You can query on-call schedules, trigger incidents, update escalation policies, analyze incident metrics, and manage service dependencies — all through natural language prompts. The MCP exposes 50 tools covering incidents, services, escalations, automation actions, and audit logs.

Do I need admin access to connect PagerDuty via OAuth?

You need a PagerDuty account with sufficient permissions to authorize the OAuth scopes Switchy requests. Typically this means Manager or Admin role, since the MCP can modify escalation policies, create incidents, and associate services. If you only have Responder access, some write operations will fail. Check your PagerDuty role before connecting.

Can the PagerDuty MCP acknowledge or resolve incidents?

Yes. The MCP includes tools to create, update, acknowledge, and resolve incidents programmatically. You can also add notes, reassign incidents, and trigger manual escalations. It doesn't replace the PagerDuty mobile app for real-time alerts, but it's useful for batch updates or building custom incident workflows inside Switchy.

How is this different from using PagerDuty's REST API directly?

The MCP wraps PagerDuty's API in natural-language tools. Instead of writing curl commands or Python scripts to aggregate escalation metrics, you ask Switchy in plain English. The trade-off: you're limited to the 50 tools Switchy exposes, so niche API endpoints might not be available. For one-off queries, the MCP is faster. For custom integrations, use the API.

Who on my team should connect the PagerDuty integration?

Whoever owns your incident response process — usually an engineering manager or SRE lead. That person's PagerDuty permissions determine what the MCP can do for the whole Switchy workspace. If multiple people need different access levels, connect separate Switchy workspaces or use PagerDuty's team-scoped API tokens outside the MCP.

Compare with

Compare with anything else →
Data last verified 7 hours ago.Sources aggregated hourly to weekly. See docs/architecture/model-directory.md.