MCP Atlassian

MCP Atlassian is an open-source Model Context Protocol (MCP) server built in Python by Hyeonsoo Lee (GitHub: sooperset) that bridges AI assistants with Atlassian's Confluence and Jira products. It supports both Atlassian Cloud and self-hosted Server/Data Center deployments, and is licensed under the MIT License. The project has accumulated over 5,600 GitHub stars and 1,200+ forks since its creation in December 2024, signaling strong community adoption.

What It Is

MCP Atlassian is a server implementation of the Model Context Protocol that exposes Jira and Confluence functionality as callable tools for AI assistants like Claude Desktop and Cursor. Rather than requiring custom API integrations, developers and teams can configure the server once and then ask their AI assistant natural-language questions like "Find issues assigned to me in PROJ project" or "Search Confluence for onboarding docs." The server handles authentication, API translation, and response formatting transparently.

Tool Coverage and Capabilities

The server ships with 98 tools total spanning both Jira and Confluence. Key tool categories include:

• Jira: jira_search (JQL), jira_get_issue, jira_create_issue, jira_update_issue, jira_transition_issue, agile board tools, comments and worklogs, links and versions, attachments, service desk, forms and metrics, and SLA metrics
• Confluence: confluence_search (CQL), confluence_get_page, confluence_create_page, confluence_update_page, confluence_add_comment, attachments, permissions, and templates

Dedicated guides for JQL (Jira Query Language) and CQL (Confluence Query Language) are included in the documentation to help users write effective queries.

Authentication and Deployment Options

MCP Atlassian supports multiple authentication methods to accommodate different Atlassian deployment types:

• API tokens for Atlassian Cloud (username + token)
• Personal Access Tokens (PATs) for Server/Data Center
• OAuth 2.0 for more advanced setups

Deployment options include running via uvx (the recommended quick-start path, described as a 5-minute setup), Docker for production environments, pip installation, or building from source. The server also supports HTTP transport modes including SSE and streamable-http for multi-user scenarios.

Compatibility

AI platform compatibility documentation is provided separately, covering IDE and assistant integrations such as Claude Desktop and Cursor.

Update: v0.23.0

The latest release is v0.23.0, published on July 18, 2026, with the repository last pushed on July 28, 2026. The project has been actively maintained since its December 2024 creation, with a full documentation site at mcp-atlassian.soomiles.com (built on Mintlify) and documentation available in llms.txt format for LLM consumption. The rapid version cadence and high star/fork counts indicate an actively developed and community-supported project.