Skip to main content
Glama

Confluence MCP Server

A Model Context Protocol server system for accessing the Confluence API. It communicates with Confluence using Personal Access Tokens. It acts as an intermediary server for accessing the Confluence API within the VPN zone.

System Configuration

This project consists of two servers:

  1. MCP Server (server.ts): The main server that communicates with the Confluence API and processes data.

    • Default port: 3000

    • Providing RESTful API

    • MCP protocol processing

  2. Proxy Server (proxyServer.ts): Acts as a proxy between the MCP client and the MCP server.

    • Default port: 3001

    • Receives requests from MCP clients and forwards them to the main server

This architecture is designed to address VPN realm issues. The MCP server runs inside the VPN and accesses the Confluence API, while the proxy server runs in an externally accessible environment.

Related MCP server: Confluence MCP

function

  • View a list of all Confluence spaces

  • View all pages within a specific space

  • Retrieve specific page information by page ID

  • Create a new page

  • Update existing page

Installation and setup

Prerequisites

  • Node.js 16.x or later

  • Bun 1.0.0 or later

installation

# 저장소 클론
git clone https://github.com/yourusername/confluence-mcp-server.git
cd confluence-mcp-server

# 의존성 설치
bun install

Setting environment variables

Create a .env file in your project root and add the following:

PORT=3000
PROXY_PORT=3001
MCP_SERVER_URL=http://localhost:3000
CONFLUENCE_BASE_URL=https://wiki.musinsa.com
PERSONAL_ACCESS_TOKEN=your_personal_access_token
NODE_ENV=development

execution

Development Environment

# MCP 서버 실행 (Confluence API와 통신)
bun run dev

# 프록시 서버 실행 (MCP 클라이언트와 통신)
bun run dev:proxy

# MCP 서버 Stdio 모드 실행 (선택 사항)
bun run dev:stdio

Production environment

# 빌드
bun run build

# MCP 서버 실행
bun run serve

# 프록시 서버 실행
bun run serve:proxy

Endpoint

MCP Server Endpoint (default port: 3000)

  • POST /mcp - Main endpoint handling MCP requests.

  • GET /mcp/tools - List available tools

  • GET /api/spaces - List all spaces

  • GET /api/spaces/:spaceKey/pages - Retrieve a list of pages in a specific space

  • GET /api/pages/:pageId - Retrieve information about a specific page

  • POST /api/pages - Create a new page

  • PUT /api/pages/:pageId - Update a page

  • GET /health - Check server health

Proxy server endpoint (default port: 3001)

  • POST /mcp - Receives MCP client request and forwards it to MCP server.

  • GET /tools - List available tools

  • GET /health - Check proxy server status

Troubleshooting

Troubleshoot VPN zone issues

This project has been split into two servers to address VPN zone issues:

  1. The MCP server runs inside the VPN and accesses the Confluence API directly.

  2. A proxy server runs in an externally accessible environment and receives requests from MCP clients.

This configuration acts as an intermediary in situations where the MCP client cannot access the Confluence API directly.

Error "Error fetching spaces: Unable to connect"

This error can be caused by the following issues:

  1. HTTPS certificate issue : SSL certificate verification is disabled in development environments, but proper certificate setup is required in production environments.

  2. Network connectivity issues : This may be due to a firewall or proxy settings preventing you from accessing the Confluence server.

  3. Personal Access Token issue : The token may have expired or be in the wrong format.

How to solve

  1. Make sure CONFLUENCE_BASE_URL in your .env file is correct.

  2. Verify that your Personal Access Token is valid.

  3. Check your firewall settings to make sure you have access to your Confluence server.

  4. Contact your network administrator to verify access rights.

  5. Make sure the connection between the MCP server and the proxy server is set up correctly (set MCP_SERVER_URL ).

test

# 모든 테스트 실행
bun test

# 특정 테스트 실행
bun test src/test/getSpaces.test.ts

Distribution Configuration

For production environments, we recommend the following deployment configuration:

  1. MCP Server : Deployed inside the VPN or on a network that has access to the Confluence server.

  2. Proxy Server : Deployed on a network accessible to MCP clients.

Security Considerations

  • We recommend that communication between the proxy server and the MCP server be encrypted using HTTPS.

  • Keep your Personal Access Token safe and renew it regularly as needed.

  • In a production environment, add an appropriate authentication mechanism to prevent unauthorized access to the proxy server.

License

MIT

A
license - permissive license
Not graded
quality - not tested
D
maintenance

Maintenance

Maintainers
Response time
Release cycle
Releases (12mo)
Commit activity

Resources

Unclaimed servers have limited discoverability.

Looking for Admin?

If you are the server author, to access and configure the admin panel.

Related MCP Servers

  • -
    license
    Not graded
    quality
    Not graded
    maintenance
    A Model Context Protocol server that enables interaction with Atlassian products (Confluence and Jira), supporting both Cloud and Server/Data Center deployments for searching, creating, and managing content through natural language.

View all related MCP servers

Related MCP Connectors

  • A comprehensive Model Context Protocol (MCP) server that enables AI assistants to interact with yo…

  • A Model Context Protocol server for Wix AI tools

  • Markdown-first MCP server for Notion API with 8 composite tools and 39 actions.

View all MCP Connectors

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/minhoyooDEV/confluence-mcp-server'

If you have feedback or need assistance with the MCP directory API, please join our Discord server