Opsgenie MCP server: community tools for alerts and on-call

Opsgenie's alerting and on-call features are moving into Jira Service Management. Atlassian's MCP server already covers that destination; small community servers cover Opsgenie itself until April 2027.

Checked 3 October 2026 · from Atlassian's Opsgenie migration page and Rovo MCP server documentation, GitHub and npm registry data

One of 91 guides to monitoring and developer tools. Each covers what an AI coding agent can get out of the tool through MCP, the protocol that lets the agent query the tool directly.

How Opsgenie works with MCP

Opsgenie routes alerts from monitoring tools to the right people, runs escalation policies and holds on-call schedules. Atlassian stopped selling it on 4 June 2025.

Atlassian is moving those capabilities into JSM Operations, the incident and on-call part of Jira Service Management. Customers have until 5 April 2027 to move their data.

After that date Atlassian says Opsgenie "will no longer be accessible" and unmoved data will be deleted. Any MCP setup against Opsgenie itself has that end date built in.

Routes from an agent to Opsgenie data

daviddykeuk/opsgenie-mcp also exists, with 1 star and a last push on 3 September 2025.

Every vendor here gets the same four checks: official, maintained, distributed and usable. Opsgenie passes 1 of them.

For a team still on Opsgenie, a community server or the REST API is the working route until April 2027. For a team that has moved to JSM Operations, Atlassian's own server is the one to keep.

What can an AI agent do with Opsgenie?

How to configure Opsgenie MCP for agents

1. 1

   Decide where the data lives

   A migrated site belongs with Atlassian's server. A site still on Opsgenie needs a community server or the REST API.

2. 2

   Choose read or write

   getJsmOpsAlerts, getJsmOpsScheduleInfo and getJsmOpsTeamInfo read. updateJsmOpsAlert changes alert state.

3. 3

   Use a service account for shared agents

   Atlassian accepts a service-account API key as a bearer token. The agent then does not depend on one person's token.

4. 4

   Check provenance on npm packages

   Two of the community packages have no reachable source repository, so their code cannot be reviewed on GitHub.

Paste this into your agent’s instructions

When you answer from Opsgenie or JSM Operations:
- Say which route the data came from: Atlassian's server, a community server or the REST API.
- Give the alert ID or schedule name you read.
If JSM tools return nothing, say the site may not have migrated from Opsgenie.
Ask before acknowledging, closing, escalating or creating any alert.

Do you need the MCP server at all?

Opsgenie's REST API remains available until shutdown. GET https://api.opsgenie.com/v2/alerts lists alerts, and the Who is On Call API returns current responders. EU accounts use api.eu.opsgenie.com.

If you are leaving Atlassian instead, PagerDuty, incident.io and Rootly each publish their own MCP servers covering incidents and on-call.

For the alerts before they reach an on-call tool, Prometheus and Grafana expose firing alerts and alert rules through their servers.

Connect what you will be running after April 2027. Treat any direct Opsgenie connection as temporary.

Opsgenie integration setup

Atlassian's server is hosted, so nothing installs. The JSM Operations tools need API-token authentication enabled by an organization admin.

claude mcp add --transport http atlassian https://mcp.atlassian.com/v2/mcp

{"mcpServers":{"atlassian-mcp":{
  "url":"https://mcp.atlassian.com/v2/mcp",
  "headers":{"Authorization":"Basic <base64-email-and-token>"}}}}

Every other client

Each block below is the configuration for one client, with the file path and the key that client expects.

curl 'https://api.opsgenie.com/v2/alerts?limit=20' \
  -H 'Authorization: GenieKey <api-key>'

What the vendor documents

Atlassian's migration page states the timeline plainly: "Opsgenie's alerting and on-call features are now available in Jira Service Management. Move existing Opsgenie data and configurations before April 5, 2027."

Should you connect an agent to Opsgenie?

Only as a bridge. Connect Atlassian's server if you have moved to JSM Operations, or a community server until April 2027. Best for teams migrating from Opsgenie to Jira Service Management who need on-call answers in the meantime.

Strongest fit

Only as a bridge. Connect Atlassian's server if you have moved to JSM Operations, or a community server until April 2027.

Main advantage

Atlassian's hosted server already covers alert search, on-call lookups and alert actions for JSM Operations.

Main weakness

Opsgenie shuts down on 5 April 2027, and its community servers are small, single-maintainer projects.

Operational risk

Alert-writing tools can acknowledge, close or escalate a real page.

Frequently asked questions