MCP Atlassian

What is MCP Atlassian?

MCP Atlassian is a Model Context Protocol server that integrates AI assistants with Atlassian Confluence and Jira. It supports both Atlassian Cloud and Server/Data Center deployments, enabling users to search, read, and modify content across these platforms through natural language commands.

How to use MCP Atlassian?

Install via uvx (recommended), pip, or Docker. Configure with command-line arguments for your Atlassian instance(s)—include only the services you need. Connect the server to an MCP client like Claude Desktop or Cursor IDE using the provided JSON or YAML configuration examples.

Key features of MCP Atlassian

• Search Confluence content using CQL
• Get, create, update, and delete Confluence pages
• Search Jira issues using JQL
• Get, create, update, and delete Jira issues
• Manage Jira worklogs and issue transitions
• Supports both Cloud and Server/Data Center deployments

Use cases of MCP Atlassian

• Ask an AI assistant to search and summarize Confluence documentation
• Create, update, or transition Jira issues directly from a chat interface
• Link issues to Epics and retrieve all issues in an Epic
• Add worklog entries to Jira issues without leaving your IDE

FAQ from MCP Atlassian

Which Atlassian products are supported?

Confluence and Jira. Both are fully supported on Cloud, and supported on Server/Data Center (Confluence 7.9+ and Jira 8.14+).

Can I use only Confluence or only Jira?

Yes. Include only the command-line arguments for the service(s) you want to use.

How do I authenticate?

For Cloud, use an API token from id.atlassian.com. For Server/Data Center, use a Personal Access Token from your profile settings.

What transport options are available?

stdio (default) and SSE. Use the --transport flag to choose SSE and --port to set the port (default 8000).

Can I run the server in read-only mode?

Yes, use the --read-only flag to disable all write operations.