SuperAPI AI Directory

confluence-mcp-server

Get Confluence data via CQL and read pages.

Open website

About this tool

Purpose

confluence-mcp-server is an MCP (Model Context Protocol) server written in TypeScript, with the goal of enabling MCP-capable clients (such as Claude Desktop) to interact with Confluence. The project describes itself as a test of a Confluence MCP server, used to demonstrate core MCP concepts.

Main capabilities

The server provides three types of tools:

  1. execute_cql_search

    • Purpose: Run CQL queries to search Confluence pages.
    • Parameters: cql, limit (default value is 10).
  2. get_page_content

    • Purpose: Retrieve the content of a Confluence page.
    • Parameters: pageId.
  3. update_page_content

    • Purpose: Update the content of a Confluence page.
    • Parameters: pageId, content, title (optional, used to modify the title).

Usage

The project provides dependency installation and build steps, and supports automatic rebuilds during development. The server can also be automatically installed for Claude Desktop via Smithery. When using it in Claude Desktop, a server configuration must be added:

  • macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
  • Windows: %APPDATA%/Claude/claude_desktop_config.json

Because MCP servers communicate via stdio, debugging may be relatively difficult; the project recommends using MCP Inspector, which is provided as a package script and, when run, gives a URL in the browser for accessing the debugging tool.

Applicable scenarios

Suitable for scenarios that require searching, reading, or updating Confluence page content in an AI client, such as retrieving documents via CQL, fetching the body by page ID, or updating the content of a specified page.

Limitations

The project description does not provide information such as the authentication method, required environment variables, specific command-line content, license, or support scope,and none is currently provided.