DeepSeek MCP server

The DeepSeek MCP Server enables seamless integration of DeepSeek's powerful language models with MCP-compatible applications like Claude Desktop, through the Model Context Protocol (MCP).

Installation Options

Via Smithery

For automatic installation with Claude Desktop using Smithery:

npx -y @smithery/cli install @dmontgomery40/deepseek-mcp-server --client claude

Manual Installation

Install globally using npm:

npm install -g deepseek-mcp-server

Configuration

Setting Up with Claude Desktop

Add the following to your claude_desktop_config.json file:

{
  "mcpServers": {
    "deepseek": {
      "command": "npx",
      "args": [
        "-y",
        "deepseek-mcp-server"
      ],
      "env": {
        "DEEPSEEK_API_KEY": "your-api-key"
      }
    }
  }
}

Usage Features

Natural Language Configuration

The server can understand and respond to natural language requests for configuration changes:

• Ask "What models are available?" to see available models
• Request "What configuration options do I have?" to view all settings
• Check "What is the current temperature setting?" to see that specific parameter
• Use commands like "Start a multi-turn conversation with deepseek-chat model, low creativity, and 8000 token limit" to configure sessions

Model Selection and Fallback

• The server defaults to deepseek-reasoner (R1 model)
• Automatic fallback to deepseek-chat (v3) if R1 is unavailable
• Switch between models anytime by adding "use deepseek-reasoner" or "use deepseek-chat" to your prompt
• V3 is recommended for general use, while R1 excels at technical and complex queries

Configurable Parameters

• Model selection (deepseek-reasoner, deepseek-chat)
• Temperature control (0.0 - 2.0)
• Maximum tokens limit
• Top P sampling (0.0 - 1.0)
• Presence penalty (-2.0 - 2.0)
• Frequency penalty (-2.0 - 2.0)

Conversation Features

Multi-turn Support

The server maintains complete conversation history and context:

• Preserves entire message history across exchanges
• Maintains configuration settings throughout conversations
• Handles complex dialogue flows automatically

This is particularly useful for:

• Training and fine-tuning custom models
• Complex problem-solving requiring context
• Multi-step reasoning tasks
• Technical discussions where previous context matters

Testing Locally

You can test the server using the MCP Inspector tool:

# Build the server first
npm run build

# Run with MCP Inspector
npx @modelcontextprotocol/inspector node ./build/index.js

The inspector interface allows you to:

• View available tools and capabilities
• Test chat completions with different parameters
• Debug server responses
• Monitor performance metrics

How to add this MCP server to Cursor

There are two ways to add an MCP server to Cursor. The most common way is to add the server globally in the ~/.cursor/mcp.json file so that it is available in all of your projects.

If you only need the server in a single project, you can add it to the project instead by creating or adding it to the .cursor/mcp.json file.

Adding an MCP server to Cursor globally

To add a global MCP server go to Cursor Settings > MCP and click "Add new global MCP server".

When you click that button the ~/.cursor/mcp.json file will be opened and you can add your server like this:

{
    "mcpServers": {
        "cursor-rules-mcp": {
            "command": "npx",
            "args": [
                "-y",
                "cursor-rules-mcp"
            ]
        }
    }
}

Adding an MCP server to a project

To add an MCP server to a project you can create a new .cursor/mcp.json file or add it to the existing one. This will look exactly the same as the global MCP server example above.

How to use the MCP server

Once the server is installed, you might need to head back to Settings > MCP and click the refresh button.

The Cursor agent will then be able to see the available tools the added MCP server has available and will call them when it needs to.

You can also explictly ask the agent to use the tool by mentioning the tool name and describing what the function does.