mcp-atlassian (community Jira/Confluence)

Add to your client

Copy the config for your MCP client and paste it into its config file.

{
  "mcpServers": {
    "mcp-atlassian-community-jira-confluence": {
      "command": "uvx",
      "args": [
        "mcp-atlassian"
      ],
      "env": {
        "JIRA_URL": "https://your-company.atlassian.net",
        "JIRA_USERNAME": "<your-email@company.com>",
        "JIRA_API_TOKEN": "<your-jira-api-token>",
        "CONFLUENCE_URL": "https://your-company.atlassian.net/wiki",
        "CONFLUENCE_USERNAME": "<your-email@company.com>",
        "CONFLUENCE_API_TOKEN": "<your-confluence-api-token>"
      }
    }
  }
}

Step-by-step guides: Add to Claude Desktop · Add to Cursor · Add to Windsurf

Before you start

• Python with `uv`/`uvx` (recommended), or Docker, to run the server; pip and from-source installs are also supported
• For Atlassian Cloud: your site URL plus an Atlassian API token (id.atlassian.com → Security → API tokens) and the account email
• For Jira/Confluence Server or Data Center: a Personal Access Token (PAT) and the base URL (Jira 8.14+, Confluence 6.0+)
• Optional: OAuth 2.0 credentials if you prefer OAuth over tokens

About mcp-atlassian (community Jira/Confluence)

Tools & capabilities (11)

What this server can do

mcp-atlassian (community Jira/Confluence) provides tools for these capabilities — tap one to see every MCP server that does the same:

When to use it

• Use it when your Jira/Confluence runs on Server or Data Center, which the official Rovo server doesn't support
• Use it when you need a local stdio server you fully control rather than a hosted remote endpoint
• Use it when you want to self-host an Atlassian MCP integration via Docker or uvx in your own environment
• Use it when an agent should search, create, update, and transition Jira issues programmatically
• Use it when reading or writing Confluence pages and comments from an AI assistant
• Use it when you need broad tool coverage (70+ tools) across both Jira and Confluence in one server

Quick setup

1. 1Install/run with `uvx mcp-atlassian` (or pull the Docker image / pip install)
2. 2Set credentials via environment variables: Cloud uses `JIRA_URL` + `JIRA_USERNAME` + `JIRA_API_TOKEN` and `CONFLUENCE_URL` + `CONFLUENCE_USERNAME` + `CONFLUENCE_API_TOKEN`; Server/DC uses `JIRA_PERSONAL_TOKEN` / PAT
3. 3Add the server to your MCP client config (e.g. Claude Desktop or Cursor) pointing at the uvx/Docker command with those env vars
4. 4Restart the client so it loads the Jira/Confluence tools
5. 5Verify by asking the assistant to run a JQL search or fetch a known Confluence page

mcp-atlassian (community Jira/Confluence) FAQ