Atlassian Confluence Data Center MCP Server (Go)

What is Atlassian Confluence Data Center MCP Server (Go)?

A Model Context Protocol (MCP) server built in Go that enables AI assistants and MCP clients to read, search, create, and update content in Atlassian Confluence Data Center edition. It uses bearer token authentication and the Confluence REST API v1.

How to use Atlassian Confluence Data Center MCP Server (Go)?

Install via pre-built binaries from the Releases page, by building from source with go build ., or using go install github.com/anudeepd/atlassian-confluence-dc-go-mcp@latest. Set the required environment variables CONFLUENCE_API_TOKEN and CONFLUENCE_BASE_URL (or CONFLUENCE_HOST / CONFLUENCE_API_BASE_PATH), then configure your MCP client (e.g., Claude Desktop) to spawn the server binary with those environment variables.

Key features of Atlassian Confluence Data Center MCP Server (Go)

• Search and retrieve content using CQL (Confluence Query Language)
• Create new pages and blog posts in Confluence spaces
• Update existing content with version management
• List and search Confluence spaces
• Bearer token authentication for secure access
• High performance with Go and minimal external dependencies

Use cases of Atlassian Confluence Data Center MCP Server (Go)

• Enable AI assistants to answer questions from enterprise Confluence documentation
• Automate the creation and updating of Confluence pages during CI/CD pipelines
• Integrate Confluence content search into custom tools and workflows
• Manage page hierarchies and space listings via programmatic access

FAQ from Atlassian Confluence Data Center MCP Server (Go)

What dependencies are required to run the server?

Go 1.25.5 or higher and network access to a Confluence Data Center instance are required.

How does authentication work?

The server uses a bearer token set via the CONFLUENCE_API_TOKEN environment variable.

Where does the data live?

All Confluence data remains in the user's own Confluence Data Center instance; the server only sends API requests to that instance and does not store any data externally.

What transport protocol does the server use?

The server communicates over stdio using the MCP protocol, compatible with any MCP client (e.g., Claude Desktop, Claude Code).

Are there any usage limits or rate limiting built in?

The README does not mention any built-in rate limiting or usage limits; refer to your Confluence Data Center instance’s own API restrictions.