atlassian

This Atlassian MCP server provides 51 MCP-compatible tools to manage your Atlassian Cloud workspace, including 13 Confluence tools for spaces, pages, and content management, and 38 Jira tools for issues, sprints, boards, and backlogs. It can be run via npx, locally installed, or deployed in Docker, with flexible authentication through environment variables, a config file, or CLI arguments. Once running, your AI assistant can invoke dedicated MCP commands to perform create, read, update, and delete operations across Confluence spaces/pages and Jira entities, as well as perform searches with CQL/JQL, manage sprints and boards, and organize backlogs and epics. The server is built with TypeScript and leverages the MCP SDK for type-safe, structured tool definitions, enabling reliable integration with AI assistants like Claude Desktop, Cursor, Claude Code, VS Code (Cline), and Windsurf.

To use the tools, start the MCP server with the chosen deployment method (npx, npm global install, or Docker) and provide authentication details for your Atlassian Cloud account. The server exposes a defined set of tools that a client can call using standard MCP conventions. For example, you can ask the AI to create a Confluence page, search for a Confluence page by CQL, create Jira issues, or move issues between sprints. The tools are grouped under Confluence (13 tools) and Jira (38 tools) and support full CRUD operations, search capabilities, and project/issue management workflows. Ensure your config includes domain, email, and API token (via config file, environment variables, or CLI) to authorize with Atlassian.

Prerequisites:\n- Node.js (for npx usage) or npm/yarn installed on your machine.\n- Access to the Atlassian Cloud instance you want to manage (domain, email, and API token).\n- Optional: Docker if you prefer containerized deployment.\n\nOption A: Run with npx (no installation)\n1. Ensure you have Node.js and npm installed.\n2. Run the MCP server directly from npm registry:\n\nbash\nnpx -y @xuandev/atlassian-mcp\n\n3. Follow the prompts or provide a config file path or environment variables as needed.\n\nOption B: Global npm installation (local dev)\n1. Install the package globally:\nbash\nnpm install -g @xuandev/atlassian-mcp\n 2. Run the server (default config file) or with a custom config:\nbash\natlassian-mcp\n# or with a custom config:\natlassian-mcp --config /path/to/config.json\n\n\nOption C: Docker deployment (containerized)\n1. Pull the image:\nbash\ndocker pull xuanxdev/atlassian-mcp:latest\n\n2. Run with a config file mounted (recommended):\nbash\ndocker run -i --rm -v ~/.atlassian-mcp.json:/config/.atlassian-mcp.json:ro xuanxdev/atlassian-mcp:latest\n 3. Run with environment variables:\nbash\ndocker run -i --rm \ -e ATLASSIAN_DOMAIN="your-domain.atlassian.net" \ -e ATLASSIAN_EMAIL="your-email@example.com" \ -e ATLASSIAN_API_TOKEN="your-api-token" \ xuanxdev/atlassian-mcp:latest\n\n\nOption D: Docker build locally (custom image)\n1. Build the image:\nbash\ndocker build -t xuanxdev/atlassian-mcp:1.0.0 -t xuanxdev/atlassian-mcp:latest .\n 2. Run with a mounted config:\nbash\ndocker run -i --rm -v /absolute/path/.atlassian-mcp.json:/config/.atlassian-mcp.json:ro xuanxdev/atlassian-mcp:latest\n

Tips and common issues:\n- Ensure your Atlassian domain, email, and API token are current and have the necessary permissions for the operations you intend to perform.\n- If using a config file, place it in a secure location with restricted permissions (chmod 600).\n- When deploying via Docker, prefer mounting a config file as a volume to keep credentials out of the container.\n- The MCP server supports multiple deployment methods; choose the one that best fits your workflow (local development vs. production).\n- If you encounter authentication errors, verify that the API token has not expired and that the domain is correct.\n- For large-scale automation, consider batching operations or using the provided search (CQL/JQL) capabilities to filter targets before performing bulk updates.\n- The environment variables approach allows easy switching between different Atlassian tenants without editing config files.