Skip to main content
mcpSource-backed

PagerDuty MCP Server for Claude

Official PagerDuty MCP Server for connecting Claude, Cursor, VS Code, and other MCP clients to PagerDuty incidents, services, schedules, on-call rotations, escalation policies, teams, event orchestrations, status pages, and incident-response workflows.

by PagerDuty · submitted by oktofeesh1·added 2026-06-04·
Review first review before installing

Open the source and read safety notes before installing.

Citation facts

Source-backed facts for citing this resource, derived directly from the registry — also available as plain text for AI assistants.

Source URLs
https://support.pagerduty.com/main/docs/pagerduty-mcp-server, https://github.com/PagerDuty/pagerduty-mcp-server, https://www.pagerduty.com/
Safety notes
PagerDuty MCP connects an AI assistant to a real incident-management account. Treat every incident, schedule, service, team, escalation policy, event orchestration, and status-page action as operationally meaningful., The local server exposes read-only tools by default. Write tools require starting the server with `--enable-write-tools`; do not enable that flag unless the workflow explicitly needs controlled writes., Write tools can create incidents, update incident status or assignment, add responders, add notes, start incident workflows, create schedule overrides, change teams, update services, edit event orchestration routing, and create status-page posts or updates., Use a least-privilege PagerDuty identity. Do not connect an all-powerful administrator token to exploratory chats, untrusted repositories, shared workstations, or broad automation loops., Require human approval before accepting assistant-proposed acknowledgements, resolutions, escalations, responder additions, schedule overrides, service updates, event-orchestration changes, or status-page communications., Tool filtering through `mcp-proxy` is documented for local subprocess setups only. For the remote server, constrain access through the PagerDuty identity, API key, OAuth client, and account permissions instead., PagerDuty's hosted remote server sends tool calls to PagerDuty-hosted MCP infrastructure. Local stdio and Docker modes run locally but still call PagerDuty APIs with the configured credential.
Privacy notes
PagerDuty MCP can expose incident titles, descriptions, notes, alerts, change events, log entries, service names, team names, user profiles, on-call schedules, escalation policies, status-page content, and event orchestration rules to the connected AI client., Incident and alert data can contain customer names, infrastructure details, internal hostnames, service dependencies, change records, remediation notes, phone or email contact data, and post-incident context., Store PagerDuty API keys, user API tokens, OAuth client secrets, and MCP config files in approved secret-management locations. Do not paste them into prompts, issues, tickets, screenshots, shell history, or committed repository files., Claude, Codex, IDE logs, MCP client transcripts, terminal history, screenshots, incident summaries, and support bundles can retain PagerDuty data outside PagerDuty's normal audit, retention, and access-control boundaries., The EU service region requires EU-specific MCP and API hosts. Verify region selection before connecting regulated incident data, customer-impacting status pages, or accounts with residency requirements.
Author
PagerDuty
Submitted by
oktofeesh1
Claim status
unclaimed
Last verified
2026-06-04

Decision playbook

Review trust signals before you adopt

Signals are present but mixed. Use the checklist below to confirm the source and operational safety for your environment.

Compare context
Selected

0

Current score

78

Baseline

Delta

No baseline selected

No major trust-signal divergence detected in the current selection.

Source and provenance checks

Complete

Confirm ownership and provenance before trusting install instructions.

  • Source link availableRequired

    Open the canonical repository and verify ownership.

    Done
  • Source provenance statusRequired

    Marked as source-backed.

    Done
  • Metadata reviewed

    Registry metadata indicates a reviewed listing.

    Done

Safety and privacy checks

Complete

Validate risk disclosures before installation or API wiring.

  • Safety notes presentRequired

    Review the listed safety guidance before running commands.

    Done
  • Privacy notes presentRequired

    Review data handling notes before connecting accounts or secrets.

    Done
  • Trust level risk gateRequired

    Trust level does not block evaluation.

    Done

Package and install checks

Needs review

Check package metadata and artifact integrity signals.

  • Install payload available

    Install or copy payload is available for review.

    Done
  • Package verification flag

    No package verification flag provided.

    Pending
  • Checksum metadata

    No checksum provided for downloaded artifact.

    Pending

Compare-driven decision checks

Needs review

Use compare context to validate trade-offs before adoption.

  • Compare tray has multiple entries

    Add at least one more entry to compare trust differences.

    Pending
  • Baseline comparison available

    No baseline peer selected yet.

    Pending
  • Diverging trust signals identified

    No major trust-signal divergence found.

    Pending

Setup at a glance

CLI install

Copy-ready — paste the snippet to get started.

15 minutes

Adoption plan

Balanced adoption plan

Current risk score 16/100. Use staged verification before broader rollout.

Risk 16

Pre-adoption checks

Validate source and review signals before any execution.

  • Confirm source provenanceRequired

    Source URL/provenance metadata is present.

    Done
  • Confirm metadata review state

    Listing has review metadata.

    Done
  • Verify install payload

    Install/config payload exists and can be inspected.

    Done

Security checks

Confirm safety, privacy, and package integrity signals.

  • Review safety notesRequired

    Safety notes are present.

    Done
  • Review privacy notesRequired

    Privacy notes are present.

    Done
  • Verify package integrity metadata

    No package verification/checksum metadata.

    Pending

Rollout

Adopt in controlled steps based on the selected plan.

  • Run in isolated sandbox firstRequired

    Use a constrained sandbox and observe behavior across multiple tasks.

    Pending
  • Roll out graduallyRequired

    Roll out to a small cohort before wider usage.

    Pending
  • Set monitoring and fallback

    Define rollback path and monitor errors after adoption.

    Pending

Evidence readiness

Evidence readiness matrix · balanced

Required evidence gates are covered (5/6 signals complete).

Risk 15

Source provenance

Present

Source repository/provenance is listed.

Required in this preset

Metadata review

Present

Review metadata is present.

Required in this preset

Safety notes

Present

Safety notes are present.

Required in this preset

Privacy notes

Present

Privacy notes are present.

Optional in this preset

Package integrity

Missing

Package integrity metadata is missing.

Optional in this preset

Install payload

Present

Install payload is available.

Required in this preset

Required evidence gates are covered for this preset.

Decision timeline

Decision timeline · balanced

5/6 steps complete with no blocking gaps for this preset.

Risk 14

triage

Confirm source provenanceRequired

Source/provenance metadata is available.

Done

triage

Check metadata review statusRequired

Review metadata is available.

Done

verify

Review safety notesRequired

Safety notes are available.

Done

verify

Review privacy notes

Privacy notes are available.

Done

verify

Validate package integrity metadata

Package integrity metadata is missing.

Pending

rollout

Verify install payload and commandsRequired

Install payload is available.

Done

No required blockers for this timeline preset.

Prerequisite readiness

Prerequisite readiness

6 prerequisites to line up before setup. Have accounts and credentials ready first.

0/6 ready
Account & credentials3Install & runtime1Network & hosting1General115 minutes

Safety & privacy surface

Safety & privacy surface

7 safety and 5 privacy notes across 4 risk areas. Review closely: credentials & tokens.

4 areas
  • SafetyGeneralPagerDuty MCP connects an AI assistant to a real incident-management account. Treat every incident, schedule, service, team, escalation policy, event orchestration, and status-page action as operationally meaningful.
  • SafetyLocal filesThe local server exposes read-only tools by default. Write tools require starting the server with `--enable-write-tools`; do not enable that flag unless the workflow explicitly needs controlled writes.
  • SafetyLocal filesWrite tools can create incidents, update incident status or assignment, add responders, add notes, start incident workflows, create schedule overrides, change teams, update services, edit event orchestration routing, and create status-page posts or updates.
  • SafetyCredentials & tokensUse a least-privilege PagerDuty identity. Do not connect an all-powerful administrator token to exploratory chats, untrusted repositories, shared workstations, or broad automation loops.
  • SafetyGeneralRequire human approval before accepting assistant-proposed acknowledgements, resolutions, escalations, responder additions, schedule overrides, service updates, event-orchestration changes, or status-page communications.
  • SafetyCredentials & tokensTool filtering through `mcp-proxy` is documented for local subprocess setups only. For the remote server, constrain access through the PagerDuty identity, API key, OAuth client, and account permissions instead.
  • SafetyCredentials & tokensPagerDuty's hosted remote server sends tool calls to PagerDuty-hosted MCP infrastructure. Local stdio and Docker modes run locally but still call PagerDuty APIs with the configured credential.
  • PrivacyLocal filesPagerDuty MCP can expose incident titles, descriptions, notes, alerts, change events, log entries, service names, team names, user profiles, on-call schedules, escalation policies, status-page content, and event orchestration rules to the connected AI client.
  • PrivacyGeneralIncident and alert data can contain customer names, infrastructure details, internal hostnames, service dependencies, change records, remediation notes, phone or email contact data, and post-incident context.
  • PrivacyCredentials & tokensStore PagerDuty API keys, user API tokens, OAuth client secrets, and MCP config files in approved secret-management locations. Do not paste them into prompts, issues, tickets, screenshots, shell history, or committed repository files.
  • PrivacyExecution & processesClaude, Codex, IDE logs, MCP client transcripts, terminal history, screenshots, incident summaries, and support bundles can retain PagerDuty data outside PagerDuty's normal audit, retention, and access-control boundaries.
  • PrivacyGeneralThe EU service region requires EU-specific MCP and API hosts. Verify region selection before connecting regulated incident data, customer-impacting status pages, or accounts with residency requirements.

Safety notes

  • PagerDuty MCP connects an AI assistant to a real incident-management account. Treat every incident, schedule, service, team, escalation policy, event orchestration, and status-page action as operationally meaningful.
  • The local server exposes read-only tools by default. Write tools require starting the server with `--enable-write-tools`; do not enable that flag unless the workflow explicitly needs controlled writes.
  • Write tools can create incidents, update incident status or assignment, add responders, add notes, start incident workflows, create schedule overrides, change teams, update services, edit event orchestration routing, and create status-page posts or updates.
  • Use a least-privilege PagerDuty identity. Do not connect an all-powerful administrator token to exploratory chats, untrusted repositories, shared workstations, or broad automation loops.
  • Require human approval before accepting assistant-proposed acknowledgements, resolutions, escalations, responder additions, schedule overrides, service updates, event-orchestration changes, or status-page communications.
  • Tool filtering through `mcp-proxy` is documented for local subprocess setups only. For the remote server, constrain access through the PagerDuty identity, API key, OAuth client, and account permissions instead.
  • PagerDuty's hosted remote server sends tool calls to PagerDuty-hosted MCP infrastructure. Local stdio and Docker modes run locally but still call PagerDuty APIs with the configured credential.

Privacy notes

  • PagerDuty MCP can expose incident titles, descriptions, notes, alerts, change events, log entries, service names, team names, user profiles, on-call schedules, escalation policies, status-page content, and event orchestration rules to the connected AI client.
  • Incident and alert data can contain customer names, infrastructure details, internal hostnames, service dependencies, change records, remediation notes, phone or email contact data, and post-incident context.
  • Store PagerDuty API keys, user API tokens, OAuth client secrets, and MCP config files in approved secret-management locations. Do not paste them into prompts, issues, tickets, screenshots, shell history, or committed repository files.
  • Claude, Codex, IDE logs, MCP client transcripts, terminal history, screenshots, incident summaries, and support bundles can retain PagerDuty data outside PagerDuty's normal audit, retention, and access-control boundaries.
  • The EU service region requires EU-specific MCP and API hosts. Verify region selection before connecting regulated incident data, customer-impacting status pages, or accounts with residency requirements.

Prerequisites

  • PagerDuty account with permission to inspect the incidents, services, schedules, teams, escalation policies, event orchestrations, and status pages the assistant should use.
  • Authentication choice: PagerDuty API key, user API token, or OAuth client depending on whether you use the hosted remote server, local stdio server, or your MCP client's authorization flow.
  • MCP-capable client such as Claude Code, Claude Desktop, Cursor, VS Code, or another client that supports local stdio or remote HTTP MCP servers.
  • For local setup, Python 3.12 and `uvx` / `uv`, or Docker when you build and run the official repository image.
  • Region choice: `https://mcp.pagerduty.com/mcp` and `https://api.pagerduty.com` for standard accounts, or `https://mcp.eu.pagerduty.com/mcp` and `https://api.eu.pagerduty.com` for EU region accounts.
  • Incident IDs, service IDs, schedule IDs, escalation policy IDs, status page IDs, team names, or time windows ready when the assistant needs precise context.

Schema details

Install type
cli
Troubleshooting
No
Source repository stats
Scope
Source repo
Collection metadata
Estimated setup
15 minutes
Difficulty
intermediate
Tool listing metadata
Full copyable content
{
  "mcpServers": {
    "pagerduty": {
      "type": "http",
      "url": "https://mcp.pagerduty.com/mcp"
    }
  }
}

About this resource

Content

PagerDuty MCP Server is PagerDuty's official Model Context Protocol server for connecting AI assistants to PagerDuty operational data. It gives Claude, Cursor, VS Code, and other MCP clients a structured way to inspect incident context, on-call schedules, services, teams, escalation policies, event orchestrations, log entries, status pages, and related incident-response resources.

Use it as an incident-response and SRE context bridge, not as an unattended automation bot. Start with the hosted remote endpoint or the local server's default read-only mode, keep the connected identity narrow, and enable write tools only for workflows where a human will review proposed operational changes before they land in PagerDuty.

Features

  • Official PagerDuty MCP Server documented by PagerDuty.
  • PagerDuty-hosted remote HTTP endpoint at https://mcp.pagerduty.com/mcp.
  • EU hosted endpoint at https://mcp.eu.pagerduty.com/mcp.
  • Local stdio server available through the official GitHub repository and uvx pagerduty-mcp.
  • Docker option for running the local server as a container.
  • Authentication support for PagerDuty API keys, user API tokens, and OAuth flows depending on the chosen server and client.
  • Read-only default for local setup, with write tools gated behind --enable-write-tools.
  • Tools for incidents, alerts, schedules, on-call assignments, services, teams, users, escalation policies, change events, log entries, incident workflows, event orchestrations, alert grouping, and status pages.
  • Tool filtering guidance for local setups through mcp-proxy.
  • Embedded MCP apps in the official repository for incident command, on-call management, compensation reporting, service dependency graphs, and onboarding experiments.

Use Cases

  • Ask Claude to summarize active incidents, affected services, notes, alerts, and recent changes before an incident handoff.
  • Check who is currently on call for a service or schedule before paging or escalating a response plan.
  • Review escalation policies, responder context, and service ownership during production triage.
  • Compare a service's incident history, related incidents, change events, and log entries while debugging a recurring outage.
  • Draft a status-page update from incident context, then require a human to review and publish it.
  • Use local tool filtering to expose only incident, on-call, or service-health tools for a focused SRE workflow.
  • Inspect event orchestration rules before proposing routing changes for noisy alerts.

Installation

Claude Code remote

For the PagerDuty-hosted remote server:

claude mcp add --transport http pagerduty https://mcp.pagerduty.com/mcp

For an EU region account:

claude mcp add --transport http pagerduty https://mcp.eu.pagerduty.com/mcp

If using an API key instead of OAuth, pass the token through your client's approved secret flow. PagerDuty's remote setup examples use a token header:

PAGERDUTY_API_KEY=<PAGERDUTY_API_KEY> claude mcp add-json pagerduty '{"type":"http","url":"https://mcp.pagerduty.com/mcp","headers":{"Authorization":"Token token=${PAGERDUTY_API_KEY}"}}'

Codex local stdio

For local stdio setup, pin the package version after checking the current PagerDuty docs and package metadata:

[mcp_servers.pagerduty-mcp-server]
command = "uvx"
args = ["pagerduty-mcp==0.17.0"]

[mcp_servers.pagerduty-mcp-server.env]
PAGERDUTY_USER_API_KEY = "<PAGERDUTY_USER_API_KEY>"
PAGERDUTY_API_HOST = "https://api.pagerduty.com"

For EU accounts, set PAGERDUTY_API_HOST to https://api.eu.pagerduty.com.

Local write tools

Only add --enable-write-tools after reviewing the write surface and approval flow:

{
  "mcpServers": {
    "pagerduty-mcp": {
      "type": "stdio",
      "command": "uvx",
      "args": ["pagerduty-mcp==0.17.0", "--enable-write-tools"],
      "env": {
        "PAGERDUTY_USER_API_KEY": "<PAGERDUTY_USER_API_KEY>",
        "PAGERDUTY_API_HOST": "https://api.pagerduty.com"
      }
    }
  }
}

Tool Surface

The current PagerDuty tools reference documents 64 tools across 14 domains: 42 read-only tools and 22 write tools. Read-only tools cover incident lookup, alert lookup, service lookup, schedules, current on-call assignments, teams, users, escalation policies, event orchestrations, change events, log entries, incident workflows, alert grouping, and status pages.

Write tools are intentionally opt-in for the local server. They include incident creation and management, responder additions, incident notes, incident workflow starts, service updates, team changes, schedule overrides, event orchestration routing changes, alert grouping changes, and status-page posts or updates.

Examples

Incident handoff

Use PagerDuty MCP in read-only mode. Summarize active incidents for service <SERVICE_ID>, including severity, current assignee, notes, recent alerts, and related change events.

On-call context

List the current on-call responders for schedule <SCHEDULE_ID> and show the next escalation policy step. Do not page or reassign anyone.

Status update draft

Draft a customer-facing status-page update for incident <INCIDENT_ID> from PagerDuty context. Show the text for approval before using any write tool.

Event orchestration review

Inspect event orchestration <ORCHESTRATION_ID> and explain which rules are likely routing noisy alerts. Do not change routing rules.

Source Notes

  • PagerDuty's knowledge-base page describes the PagerDuty MCP Server as an open standard that lets AI assistants such as Claude Desktop, Cursor, and VS Code interact securely with a PagerDuty account.
  • PagerDuty documents two integration options: a PagerDuty-hosted service at mcp.pagerduty.com/mcp and a self-hosted open-source implementation.
  • PagerDuty documents EU hosted configuration at https://mcp.eu.pagerduty.com/mcp and EU API host configuration through https://api.eu.pagerduty.com.
  • The official developer docs describe remote setup with API-key and OAuth authorization, Cursor, VS Code, and Claude examples, and no local Python or uvx requirement for the hosted remote server.
  • The official tools reference documents 64 tools across 14 domains, with 42 read-only tools available by default and 22 write tools requiring --enable-write-tools for local setup.
  • PagerDuty's official GitHub repository describes the local MCP server, client setup, Docker setup, read-only defaults, write-tool flag, embedded MCP apps, and support through GitHub issues or the PagerDuty community forum.
  • The PyPI package metadata checked at submission time showed pagerduty-mcp==0.17.0, package name io.github.PagerDuty/pagerduty-mcp, and Python requirement ~=3.12.0.

Duplicate Check

Checked current upstream/main, open PR titles, open PR changed files, source URLs, and content files for PagerDuty, PagerDuty MCP, pagerduty-mcp-server, pagerduty-mcp, mcp.pagerduty.com, mcp.eu.pagerduty.com, api.pagerduty.com, PagerDuty/pagerduty-mcp-server, and support.pagerduty.com/main/docs/pagerduty-mcp-server. No dedicated PagerDuty MCP Server entry, PagerDuty source URL duplicate, or open content PR for this server was found.

Editorial Disclosure

PagerDuty is a commercial incident-management and operations platform, but this listing is not sponsored, paid, affiliate-backed, or submitted by PagerDuty. Use PagerDuty's current docs, package metadata, account permissions, region requirements, privacy terms, and incident-management policies as the source of truth before connecting operational data to any AI client.

Source citations

Add this badge to your README

Show that PagerDuty MCP Server for Claude is listed on HeyClaude. Paste this Markdown into your README — it renders the badge and links back to this page.

Listed on HeyClaude
[![Listed on HeyClaude](https://heyclau.de/badge/mcp/pagerduty-mcp-server.svg)](https://heyclau.de/entry/mcp/pagerduty-mcp-server)

How it compares

PagerDuty MCP Server for Claude side by side with 3 alternatives on trust, install, platform support, and disclosed safety notes — all from reviewed registry metadata.

1 trust signal differ across this comparison (Submitter).

Field

Official PagerDuty MCP Server for connecting Claude, Cursor, VS Code, and other MCP clients to PagerDuty incidents, services, schedules, on-call rotations, escalation policies, teams, event orchestrations, status pages, and incident-response workflows.

Open dossier

Manage production incidents from Claude — create incidents, check on-call schedules, find related past incidents, suggest solutions, track post-mortems, and orchestrate response workflows — with the official Rootly remote MCP server.

Open dossier

Official New Relic AI MCP Server for connecting Claude, Codex, VS Code, Windsurf, Gemini CLI, and other MCP clients to New Relic observability data, NRQL, alerts, incidents, entities, logs, performance analysis, and deployment impact context.

Open dossier

Remote Agentage Memory MCP server giving Claude one shared, portable memory across every AI you use - search, read, write, edit, list, and delete plain Markdown notes you own, over Streamable HTTP with OAuth 2.1.

Open dossier
Next steps
Trust
Review statusReviewedMaintainer reviewedReviewedMaintainer reviewedReviewedMaintainer reviewedReviewedMaintainer reviewed
Package trustPackage not verifiedPackage not verifiedPackage not verifiedPackage not verified
Source provenanceSource-backedSource-backedSource-backedSource-backed
SubmitterDiffersoktofeesh1oktofeesh1vreshch
Install riskReview firstReview firstReview firstReview first
Notes Safety ✓ Privacy ✓ Safety ✓ Privacy ✓ Safety ✓ Privacy ✓ Safety ✓ Privacy ✓
BrandAgentage Memory logoAgentage Memory
Categorymcpmcpmcpmcp
SourceSource-backedSource-backedSource-backedSource-backed
AuthorPagerDutyRootly AI LabsNew Relicagentage
Added2026-06-042026-06-182026-06-042026-06-18
Platforms
Harness
Source repo
Safety notesPagerDuty MCP connects an AI assistant to a real incident-management account. Treat every incident, schedule, service, team, escalation policy, event orchestration, and status-page action as operationally meaningful. The local server exposes read-only tools by default. Write tools require starting the server with `--enable-write-tools`; do not enable that flag unless the workflow explicitly needs controlled writes. Write tools can create incidents, update incident status or assignment, add responders, add notes, start incident workflows, create schedule overrides, change teams, update services, edit event orchestration routing, and create status-page posts or updates. Use a least-privilege PagerDuty identity. Do not connect an all-powerful administrator token to exploratory chats, untrusted repositories, shared workstations, or broad automation loops. Require human approval before accepting assistant-proposed acknowledgements, resolutions, escalations, responder additions, schedule overrides, service updates, event-orchestration changes, or status-page communications. Tool filtering through `mcp-proxy` is documented for local subprocess setups only. For the remote server, constrain access through the PagerDuty identity, API key, OAuth client, and account permissions instead. PagerDuty's hosted remote server sends tool calls to PagerDuty-hosted MCP infrastructure. Local stdio and Docker modes run locally but still call PagerDuty APIs with the configured credential.The MCP server can create and update production incidents — review Claude's proposed actions before confirming write operations during live incidents. Security-sensitive operations and bulk deletions are excluded from the default tool set; the hosted endpoint returns approximately 70 frequently-used tools in 'slim' mode. A Global API Key is recommended for full cross-team visibility; team-scoped keys may return incomplete results for on-call and analytics tools.New Relic documents the MCP Server as a preview feature. Re-check the official overview, setup guide, and tool reference before relying on a remembered endpoint, role requirement, authentication flow, tool tag, or tool name. New Relic warns that using AI tools with its MCP servers means the AI can take action on the user's behalf. Use a least-privilege New Relic account and keep assistant work read-only unless a human has explicitly approved the operation. Tool calls are governed by the permissions of the configured New Relic user API key or OAuth profile. RBAC failures are safer than over-broad access; do not use admin-like credentials just to make the assistant easier to use. The tool reference includes NRQL execution, natural-language-to-NRQL, alert and incident queries, entity discovery, dashboard access, logs, and deployment impact analysis. Treat generated queries and summaries as drafts until an operator checks the account, time window, filters, and service. Use `include-tags` to reduce the tool corpus when the task is narrow. For example, an alert triage session may only need `discovery,alerting` rather than every data-access, incident-response, performance, and advanced analysis tool. Do not commit API keys, OAuth tokens, `mcp.json`, Claude Desktop configs, Codex configs, or copied New Relic headers that contain real credentials.The memory__write, memory__edit, and memory__delete tools create, modify, and remove your stored notes; review actions before granting broad access.
Privacy notesPagerDuty MCP can expose incident titles, descriptions, notes, alerts, change events, log entries, service names, team names, user profiles, on-call schedules, escalation policies, status-page content, and event orchestration rules to the connected AI client. Incident and alert data can contain customer names, infrastructure details, internal hostnames, service dependencies, change records, remediation notes, phone or email contact data, and post-incident context. Store PagerDuty API keys, user API tokens, OAuth client secrets, and MCP config files in approved secret-management locations. Do not paste them into prompts, issues, tickets, screenshots, shell history, or committed repository files. Claude, Codex, IDE logs, MCP client transcripts, terminal history, screenshots, incident summaries, and support bundles can retain PagerDuty data outside PagerDuty's normal audit, retention, and access-control boundaries. The EU service region requires EU-specific MCP and API hosts. Verify region selection before connecting regulated incident data, customer-impacting status pages, or accounts with residency requirements.Incident details, on-call schedules, escalation policies, team configurations, and post-incident review data from your Rootly workspace are surfaced in Claude's context. OAuth is the recommended authentication method — no API token is stored in your MCP configuration.Tool results can expose New Relic account IDs, entity GUIDs, dashboard names, alert policy details, issue and incident state, NRQL query text, NRDB results, logs, error groups, change events, deployment markers, service names, tags, thread metrics, Kafka metrics, and other production observability data. Natural-language prompts can be converted into NRQL and executed against New Relic data. Review prompts, generated queries, and query results before pasting them into tickets, chat systems, public issues, or vendor support threads. Claude, Codex, IDE transcripts, MCP client logs, screenshots, exported incident reports, and shell history may retain New Relic-derived context outside New Relic's normal access controls. OAuth tokens and user API keys connect the assistant to a real New Relic identity. Rotate exposed keys, remove stale client configs, and audit account membership when a project or teammate no longer needs access.Notes you read or write are stored server-side in your Agentage Memory and mirrored locally as Markdown you can export anytime; note contents pass through model context.
Prerequisites
  • PagerDuty account with permission to inspect the incidents, services, schedules, teams, escalation policies, event orchestrations, and status pages the assistant should use.
  • Authentication choice: PagerDuty API key, user API token, or OAuth client depending on whether you use the hosted remote server, local stdio server, or your MCP client's authorization flow.
  • MCP-capable client such as Claude Code, Claude Desktop, Cursor, VS Code, or another client that supports local stdio or remote HTTP MCP servers.
  • For local setup, Python 3.12 and `uvx` / `uv`, or Docker when you build and run the official repository image.
  • A Rootly account — authenticate via OAuth when prompted, or generate an API key in Rootly → Account → Manage API Keys.
  • An MCP client such as Claude Code or Claude Desktop.
  • New Relic account with API access and the New Relic AI MCP Server public preview enabled.
  • New Relic group membership with an organizational role documented for MCP access, such as Organization Read Only, Organization Manager, or Organization Product Admin.
  • OAuth authentication, or a New Relic user API key approved for the target account and stored outside committed project files.
  • MCP-capable client such as Claude Code, Claude Desktop, Codex CLI, VS Code, Windsurf, Gemini CLI, or another compatible environment.
  • agentage account (sign in at memory.agentage.io)
  • OAuth 2.1 authentication (PKCE + Dynamic Client Registration, handled in the connect flow)
  • Network access to memory.agentage.io (HTTPS required)
  • Claude Desktop 0.7.0+ or Claude Code with MCP (HTTP transport) support
Install
claude mcp add --transport http pagerduty https://mcp.pagerduty.com/mcp
claude mcp add --transport http rootly https://mcp.rootly.com/mcp
claude mcp add newrelic --transport http https://mcp.newrelic.com/mcp/
claude mcp add --transport http agentage-memory https://memory.agentage.io/mcp && claude mcp list
Config
{
  "mcpServers": {
    "pagerduty": {
      "type": "http",
      "url": "https://mcp.pagerduty.com/mcp"
    }
  }
}
{
  "mcpServers": {
    "rootly": {
      "type": "http",
      "url": "https://mcp.rootly.com/mcp"
    }
  }
}
{
  "mcpServers": {
    "newrelic": {
      "type": "http",
      "url": "https://mcp.newrelic.com/mcp/",
      "headers": {
        "include-tags": "discovery,alerting"
      }
    }
  }
}
{
  "mcpServers": {
    "agentage-memory": {
      "url": "https://memory.agentage.io/mcp",
      "type": "http"
    }
  }
}
Citations
ClaimUnclaimedUnclaimedUnclaimedUnclaimed
Open 4 picks in the interactive comparison tool

Related guides

Signals

Loading live community signals…

More like this, weekly

A short, calm digest of reviewed Claude resources. Unsubscribe any time.