#
Atlassian Rovo MCP Server
#
Overview
The Atlassian Rovo MCP Server brings your team's Atlassian Cloud workspace into the conversation. Through the Model Context Protocol (MCP), it gives AI-powered applications structured, permissioned access to Jira, Confluence, Jira Service Management, and Compass — so the assistant can read, reason over, and act on the same data your teams work with every day.
Crucially, every interaction respects the permissions, roles, and workspace boundaries already configured in Atlassian. Users see and act on only what they're authorised to — meaning AI workflows inherit your existing governance model rather than working around it. This makes the Rovo MCP Server suitable for environments where compliance, auditability, and access control aren't optional.
#
What it unlocks in Phoeniqs Chat
When connected to Phoeniqs Chat, the Rovo MCP Server turns natural-language prompts into Atlassian actions. Common use cases include:
- Project visibility — query Jira issues, epics, and sprint status without leaving the chat
- Knowledge retrieval — summarise Confluence pages in seconds
- Ticket operations — create, update, and comment on issues directly from a conversation
- Service support — assist with incident triage and service request handling
The result: less context-switching, faster collaboration, and Atlassian data that becomes actionable — not just searchable — while preserving the security and traceability of Atlassian's cloud platform.
#
Connecting your MCP client to the Rovo MCP Server
This section gives a high-level view of the connection flow. For client-specific instructions, follow Atlassian's official documentation:
Use the procedure below to understand the overall flow before working through Atlassian's detailed steps.
#
Procedure
1. Review the official setup guide Read Atlassian's getting-started documentation to understand prerequisites, authentication, supported clients, and workflow examples.
2. Verify prerequisites Confirm the following are in place:
- An Atlassian Cloud site with access to Jira, Confluence, and/or Compass
- A supported MCP client — for example, an AI assistant, IDE, or custom tool
- A modern browser for the OAuth authorisation flow
- For IDE or desktop proxy setups: Node.js v18 or higher (if using a local proxy)
3. Authenticate and authorise Complete Atlassian's OAuth consent flow. Access is granted based on the authenticated Atlassian user's permissions, ensuring the MCP client inherits — not overrides — that user's access scope.
4. Configure your MCP client Follow Atlassian's client-specific instructions (for example, VS Code or Cursor) to point your MCP client at the Atlassian Remote MCP Server.
5. Verify the connection Run a simple test — querying or summarising Atlassian content — to confirm the client is connected and authorised end-to-end.
#
Security and data sovereignty
This integration connects to Atlassian Cloud via Atlassian's Remote MCP Server. Actions and data access are governed by:
- Atlassian's security model — TLS in transit and OAuth for authentication
- The authenticated user's permissions — every action is scoped to what that user can already see and do
⚠️ Sovereignty note: if your organisation has data residency or regulatory requirements that restrict cloud tooling, review this integration with your security and compliance teams before enabling it.