Skip to directory

About Atlassian Rovo MCP

A cloud-hosted MCP server that connects AI clients to Jira, Confluence, Jira Service Management, Bitbucket, Loom, and other Atlassian Cloud data through Teamwork Graph.

What you can use it for

  • Search and summarize Jira work items and Confluence content
  • Create or update work items and pages
  • Retrieve context from connected Atlassian apps and Teamwork Graph
Connection and setup
Hosting
Remote, cloud-hosted
Transport
Streamable HTTP
Endpoint
https://mcp.atlassian.com/v2/mcp
Authentication
OAuth 2.1 interactive consent. Organization-admin-enabled API token authentication is also documented for machine-to-machine use; some tool groups require one method specifically.

Requirements

  • An Atlassian Cloud site with at least one supported product
  • A compatible MCP client
  • A browser to complete OAuth login; API token use requires organization-admin enablement

Setup steps

  1. Add the remote endpoint https://mcp.atlassian.com/v2/mcp in a supported MCP client.
  2. Start the server authentication flow and sign in to Atlassian in the browser.
  3. Approve the requested access and verify available tools in the client. Access follows the signed-in user's Atlassian permissions.
Tools 16+

The full catalog is dynamic and depends on user scopes, product access, organization settings, and authentication method. Stored rows are the documented primary tools plus representative named tools, not the entire session catalog. The Oct 28, 2026 changelog promotes listJiraIssueComments to a primary tool; discover and execute tools expose further deferred capabilities.

  • atlassianUserInfo

    Retrieve the authenticated user's Atlassian account information.

    01
  • getAccessibleAtlassianResources

    List accessible Atlassian sites and apps with cloud IDs.

    02
  • discover

    Find deferred tools using a natural-language request.

    03
  • executeRead

    Run a discovered read operation.

    04
  • executeWrite

    Run a discovered write operation.

    05
  • executeDestructive

    Run a discovered destructive operation.

    06
  • searchJiraIssuesUsingJql

    Search Jira work items using JQL.

    07
  • getJiraIssue

    Get a Jira work item by ID or key.

    08
  • createJiraIssue

    Create a Jira work item.

    09
  • getConfluenceContent

    Read Confluence pages and other supported content.

    10
  • searchConfluence

    Search Confluence content with CQL.

    11
  • createConfluenceContent

    Create supported Confluence content.

    12
  • getTeamworkGraphContext

    Retrieve connected context and relationships for Atlassian entities.

    13
  • search

    Semantic search across Jira, Confluence, and connected apps (beta).

    14
  • getLoomVideo

    Retrieve Loom video metadata and optionally transcript or related details.

    15
  • listJiraIssueComments

    List comments on a Jira issue.

    16

This server’s available tools can vary by account, permissions, or configuration. The list shown is not its full runtime catalog.

Limitations and notes
  • The tools available in a session vary by authentication method, OAuth scopes, admin-enabled permission groups, and the user's app access.
  • Atlassian says some calls can consume Rovo credits; usage depends on tool and context.
Source discrepancies
  • The MCP.Directory profile described the implementation as local stdio; Atlassian's current documentation describes the official Rovo server as cloud-hosted at the remote v2 endpoint. This record follows Atlassian's primary documentation.
Original sources
  • official tool reference

    Reviewed current publisher or maintainer documentation for the tool inventory, availability, endpoint, or configuration.

  • official documentation

    Reviewed current publisher or maintainer documentation for the tool inventory, availability, endpoint, or configuration.

Documentation checked 2026-10-28T00:00:00Z. Connection has not been runtime tested.

Similar tools