Confluence MCP Server

Confluence MCP Server connects Atlassian Confluence to AI assistants through the Model Context Protocol, enabling LLMs to search, read, create, and update Confluence pages without leaving their AI client. The most widely used implementation is mcp-atlassian, which supports both Confluence Cloud and Server/Data Center editions (v6.0+) with 72 tools total across Jira and Confluence. Confluence-specific tools include searching with CQL (Confluence Query Language), retrieving page content with ancestors and descendants, listing spaces, creating and updating pages, and managing both footer and inline comments. Content is automatically converted to Markdown format for clean LLM consumption. Atlassian also offers an official Remote MCP Server hosted on Cloudflare that covers both Jira and Confluence Cloud with OAuth 2.1 authentication. The community mcp-atlassian server connects via API token for Cloud or personal access token for Server/Data Center, supporting both transport types. Common use cases include pulling documentation context into coding sessions, generating meeting notes as Confluence pages, searching knowledge bases through natural language, and keeping documentation synchronized with project changes. The server works with Claude Desktop, Cursor, VS Code, and other MCP-compatible clients.