mcp-server-dify

View original on GitHub

Model Context Protocol Server for Dify AI. This server enables LLMs to interact with Dify AI’s chat completion capabilities through a standardized protocol.

Features

Integration with Dify AI chat completion API
Restaurant recommendation tool (meshi-doko)
Support for conversation context
Streaming response support
TypeScript implementation

Installation

Using NPM


npm install @modelcontextprotocol/server-dify

Usage

With Claude Desktop

Add the following configuration to your claude_desktop_config.json:


{
  "mcpServers": {
    "dify": {
      "command": "npx",
      "args": [
        "-y",
        "@modelcontextprotocol/server-dify",
        "https://your-dify-api-endpoint",
        "your-dify-api-key"
      ]
    }
  }
}

Replace your-dify-api-endpoint and your-dify-api-key with your actual Dify API credentials.

Tools

meshi-doko

Restaurant recommendation tool that interfaces with Dify AI:

Parameters:

LOCATION (string): Location of the restaurant
BUDGET (string): Budget constraints
query (string): Query to send to Dify AI
conversation_id (string, optional): For maintaining chat context

Development

This project uses a monorepo structure with workspaces. The main server implementation is in the src/dify directory.

Setup


# Install dependencies
npm install
 
# Build all packages
npm run build
 
# Development mode with watch
npm run watch
 
# Run linter
npm run lint
 
# Fix linting issues
npm run lint:fix

Publishing


# Publish all packages
npm run publish-all

Security

This server interacts with Dify AI using your provided API key. Ensure to:

Keep your API credentials secure
Use HTTPS for the API endpoint
Never commit API keys to version control

License

MIT License - see LICENSE file for details.

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.