Mural MCP Server
Enables AI assistants to interact with Mural visual collaboration workspaces through OAuth 2.0 authentication, allowing users to list and retrieve workspace information.
README
Mural MCP Server
A Model Context Protocol (MCP) server that provides integration with the Mural visual collaboration platform. This server enables AI assistants to interact with Mural workspaces through OAuth 2.0 authentication.
Note: This is v0.0.1 - an early MVP release focused on workspace listing functionality. More features are planned for future releases.
Features
- OAuth 2.0 Authentication: Secure authentication with PKCE support
- Workspace Management: List and retrieve workspace information
- MCP Compliance: Full Model Context Protocol compatibility
- Token Management: Automatic token refresh and secure storage
Tools Available
list-workspaces: List all workspaces the authenticated user has access toget-workspace: Get detailed information about a specific workspacetest-connection: Test the connection to Mural API and verify authenticationclear-auth: Clear stored authentication tokens
Prerequisites
- Node.js: Version 18 or higher
- Mural Account: Access to Mural with workspace permissions
- Mural OAuth App: Register an app at Mural Developer Portal
Installation
Option 1: Install from npm (Recommended)
npm install -g mural-mcp
# or
pnpm add -g mural-mcp
Option 2: Install from source
- Clone and install dependencies:
git clone https://github.com/your-username/mural-mcp.git
cd mural-mcp
pnpm install
- Build the project:
pnpm run build
Setup
- Set up environment variables:
cp .env.example .env
# Edit .env with your Mural OAuth credentials
Configuration
Environment Variables
Create a .env file with the following variables:
# Required: Your Mural app's client ID
MURAL_CLIENT_ID=your_client_id_here
# Optional: Your Mural app's client secret (recommended)
MURAL_CLIENT_SECRET=your_client_secret_here
# Optional: OAuth redirect URI (defaults to http://localhost:3000/callback)
MURAL_REDIRECT_URI=http://localhost:3000/callback
Mural OAuth App Setup
- Visit Mural Developer Portal
- Create a new application
- Set the redirect URI to
http://localhost:3000/callback(or your custom URI) - Note your Client ID and Client Secret
- Configure the required scopes:
workspaces:read
Usage
With Claude Desktop
Add to your Claude Desktop configuration (claude_desktop_config.json):
{
"mcpServers": {
"mural": {
"command": "node",
"args": ["/absolute/path/to/mural-mcp/build/index.js"],
"env": {
"MURAL_CLIENT_ID": "your_client_id_here",
"MURAL_CLIENT_SECRET": "your_client_secret_here"
}
}
}
}
Standalone Usage
Run the server directly:
# Set environment variables
export MURAL_CLIENT_ID=your_client_id_here
export MURAL_CLIENT_SECRET=your_client_secret_here
# Start the server
pnpm start
Development
For development with hot reloading:
pnpm run dev
Authentication Flow
- First Run: The server will open a browser window for OAuth authentication
- Login: Complete the OAuth flow in your browser
- Token Storage: Tokens are securely stored locally in
~/.mural-mcp-tokens.json - Auto-Refresh: Access tokens are automatically refreshed when needed
API Endpoints Used
- Authorization:
https://app.mural.co/api/public/v1/authorization/oauth2/ - Token Exchange:
https://app.mural.co/api/public/v1/authorization/oauth2/token - Workspaces:
https://app.mural.co/api/public/v1/workspaces
Example Usage
Once configured, you can use the tools through your MCP client:
Human: List my Mural workspaces
Assistant: I'll list your Mural workspaces using the list-workspaces tool.
[Tool executes and returns workspace data]
Security Considerations
- PKCE: Uses Proof Key for Code Exchange for enhanced OAuth security
- Token Storage: Tokens are stored locally in the user's home directory
- HTTPS: All API communications use HTTPS
- Scope Limitation: Requests only necessary OAuth scopes
Troubleshooting
Authentication Issues
- Token Expired: Use the
clear-authtool to clear tokens and re-authenticate - Invalid Client ID: Verify your
MURAL_CLIENT_IDin the environment variables - Redirect URI Mismatch: Ensure the redirect URI matches your Mural app configuration
Connection Issues
- Network: Ensure you can reach
https://app.mural.co - Firewall: Port 3000 must be available for OAuth callback
- Test Connection: Use the
test-connectiontool to verify API access
Common Error Messages
Missing required environment variable: MURAL_CLIENT_ID: Set the required environment variableOAuth token exchange failed: Check your client credentials and redirect URIMural API request failed: HTTP 401: Token expired or invalid, clear auth and re-authenticateMural API request failed: HTTP 403: Insufficient permissions or invalid scope
Development
Project Structure
mural-mcp/
├── src/
│ ├── index.ts # Main MCP server
│ ├── oauth.ts # OAuth 2.0 implementation
│ ├── mural-client.ts # Mural API client
│ └── types.ts # TypeScript interfaces
├── build/ # Compiled output
├── spec/ # Documentation
└── package.json
Building
# Clean build
rm -rf build && pnpm run build
# Development build with watch
pnpm run dev
Testing
Test the server manually:
# Build and run
pnpm run build
node build/index.js
# In another terminal, test with MCP inspector
npx @modelcontextprotocol/inspector node build/index.js
Contributing
- Fork the repository
- Create a feature branch
- Make your changes
- Add tests if applicable
- Submit a pull request
License
MIT License - see LICENSE file for details
Support
For issues and questions:
- Check the troubleshooting section above
- Review Mural API documentation
- Create an issue in the repository
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。