Integrate Jira and Confluence with AI

An MCP server for Atlassian products - Jira and Confluence - supporting both Cloud and Server/Data Center deployments, with 72 tools total. Key Jira tools include jira_search (search with JQL), jira_get_issue (get issue details), jira_create_issue, jira_update_issue, and jira_transition_issue (change status). Key Confluence tools include confluence_search (search with CQL), confluence_get_page, confluence_create_page, confluence_update_page, and confluence_add_comment. This lets an AI assistant find issues assigned to a user in a specific project, search Confluence for documentation, create a bug ticket, or update an issue's status conversationally, across the full 72-tool surface documented in the project's Tools Reference.

Use it when an AI assistant needs to search, read, create, or update Jira issues or Confluence pages - triaging assigned work, looking up documentation, filing bugs, or transitioning issue status. Confluence Cloud and Jira Cloud are fully supported; Confluence Server/Data Center is supported from v6.0+ and Jira Server/Data Center from v8.14+, using a Personal Access Token (JIRA_PERSONAL_TOKEN) instead of a username plus API token for those on-premises deployments.

Input is a natural-language request (a JQL/CQL-style search, an issue key, or the fields for a new issue or page). Output is matching issues or pages, a created/updated issue or page confirmation, or a structured error when a request fails - the project ships a dedicated Troubleshooting doc page covering common issues and debugging steps for exactly these failure cases.

{
  "mcpServers": {
    "mcp-atlassian": {
      "command": "uvx",
      "args": ["mcp-atlassian"],
      "env": {
        "JIRA_URL": "https://your-company.atlassian.net",
        "JIRA_USERNAME": "your.email@company.com",
        "JIRA_API_TOKEN": "your_api_token",
        "CONFLUENCE_URL": "https://your-company.atlassian.net/wiki",
        "CONFLUENCE_USERNAME": "your.email@company.com",
        "CONFLUENCE_API_TOKEN": "your_api_token"
      }
    }
  }
}

Installable via uvx with no separate clone step, or via Docker, pip, or from source for teams that prefer a pinned local build. Get an API token from id.atlassian.com's security settings (Cloud) or use a Personal Access Token (Server/Data Center). Full documentation, including an llms.txt-format sitemap and complete documentation file for LLM consumption, is published at mcp-atlassian.soomiles.com, alongside dedicated pages on HTTP transport options (SSE, streamable-http, multi-user) for teams running it as a shared service rather than a per-user local process.

Teams using Jira and/or Confluence who want an AI assistant to search, read, create, and update issues and pages conversationally across either Atlassian Cloud or an on-premises Server/Data Center deployment, without constantly switching between the separate Jira and Confluence web interfaces to look things up. API tokens and .env files should never be shared and should be kept secure, per the project's own security guidance. Distributed under the MIT License; not an official Atlassian product.