Jira Service Management is live on Atlassian MCP: Service context and action to where you work
Service and operations work depends on context. What issues customers are running into, which requests need attention before breaching their SLA, or who’s on-call to handle a growing set of alerts.
That work doesn’t just happen in your ITSM platform anymore. It may start in Slack or Teams, an AI assistant like Claude, or another application entirely. The bottom line: you need crucial context to be available where you work so you aren’t wasting time switching between multiple tabs.
As part of the major upgrade to the Atlassian MCP, a host of new tools are available for Jira Service Management customers. MCP-compatible clients and agents can use supported tools to securely search, read, create, and update service and operations work to do things like:
- Create, triage, and resolve requests
- Acknowledge, close, and escalate alerts
- View on-call schedules and availability
- Search for assets
- Draft new knowledge
Powered by the Teamwork Graph, AI workflows with Atlassian MCP are grounded in the unique organizational knowledge across all your apps, not just Jira Service Management. With centralized admin controls and OAuth, Atlassian MCP respects your existing Atlassian permissions, an AI agent only sees and changes what you are allowed to access.
While Atlassian MCP helps you stay in your preferred flow of work, the work you do inside Atlassian MCP syncs back to Jira Service Management. With a compounding system of record, you continuously strengthen the Teamwork Graph, your organization’s unique context, and your overall Atlassian product experience while having a clear trail for audit and compliance.
What can you do with it?
Here’s a few ways you can work inside your AI tools.
Surface requests that need your attention
Find open requests in the IT service project that are approaching or have breached their SLA. Highlight any available Request Resolver plans, including the status, proposed steps, and linked runbooks from Confluence. Then, recommend what I should work on first.
Instead of manually sifting through tickets, the service agent can know exactly which requests need their input, the customer sentiment, and review Request Resolver plans (our L1 resolution agent) to help customers fast.
Set up a new starter with the right access
Create laptop, email, and Slack access requests for a new starter who begins Monday. Include the required details for each request so they can be completed without follow-up.
Raise the access and equipment requests a new starter needs, with the information teams need to act on them.
Search for the right knowledge
Review this VPN support request and find the relevant troubleshooting guide in Confluence. Compare the guidance with the information in the request, flag anything missing, and draft a response to guide the customer.
Help a service agent understand a request, find the documented guidance in Confluence, and prepare an informed response.
Learn more about a specific asset
Review this laptop support request, find the laptop in Assets, and summarize its recorded model, status, and warranty details. Compare those details with our laptop replacement policy in Confluence and recommend whether the next step is troubleshooting, repair, or replacement.
Keep the service agent agent from having to locate the asset record, copy its details, find the relevant policies, and more in order to help the customer accordingly.
Move approvals forward
Summarize all requests waiting for my approval, including the requester, what they need, and any relevant context. Recommend whether I should approve or decline each request.
Help a service agent identify the approvals holding up work and make informed decisions without leaving their flow.
Review Request Resolver plans safely
Show the Request Resolver plan for this request, including its status, proposed steps, and linked Confluence runbooks. Highlight any risks or missing information before I approve the plan.
Bring Request Resolver (our L1 resolution agent) plans into an AI workflow so a service agent can review the proposed resolution and its supporting guidance before approving it.
Investigate and escalate alerts
Find active Jira Service Management alerts. Summarize the issue and priority, who’s currently on call to respond, and the escalation policies for their teams. Recommend next best steps before I choose to acknowledge or escalate the alert.
A service agent can spend more time responding to the alert than gathering background context like on the alert itself, who is on-call, or the relevant escalation policy.
📖 Explore the full list of available Jira Service Management tools, as well as other Atlassian apps, in the Atlassian MCP supported tools documentation.
How to get started
You’ll need access to Atlassian MCP and a compatible AI client.
- Confirm your organization admin has made MCP available for Jira Service Management.
- Connect your client. Most AI tools can install, authenticate, and set up Atlassian MCP from a prompt. Copy and paste this:
- Set up Atlassian MCP for this agent using the official setup guide at https://support.atlassian.com/atlassian-ai-gateway/docs/get-started-with-the-atlassian-remote-mcp-server/ and the MCP server URL https://mcp.atlassian.com/v2/mcp. Then start the Atlassian MCP authentication flow so I can sign in.
- Authenticate your account. On the first connection, you will need to authorize with your Atlassian account. With secure OAuth 2.1 authorization, your existing access controls and permissions are respected.
- Already connected to Atlassian MCP? Reauthorize to grant the new JSM scopes, then ask your AI assistant to use the JSM tools.
- Start with a simple prompt like “Who is currently on-call” to confirm that the connection was successful.
Refer to Atlassian MCP documentation for more help with set up.
If you were using Atlassian MCP before the recent upgrade, follow these instructions to upgrade to the latest version.
We’d love to hear from you
We’d love to hear how you’re planning to put the Jira Service Management tools in Atlassian MCP to work. Drop us a comment below on what use cases have become a part of your daily workflow.
As for what’s next, we’re exploring how existing Jira Service Management Slack and Microsoft Teams experiences can become a more complete system of engagement —bringing service context and action into the flow of conversation, not only creating requests.
➡️ Book time with our team if you’d love to provide feedback so we continue to build with customers in mind.
Juhi Duseja
PM, JSM | Atlassian