DeepSeek Thinker MCP server

Deepseek Thinker MCP is a server that provides access to Deepseek's reasoning capabilities through the Model Context Protocol (MCP). It allows AI clients like Claude Desktop to leverage Deepseek's thought processes either via the Deepseek API service or through a local Ollama server.

Features

• Dual Mode Support: Works with either OpenAI API or local Ollama installation
• Focused Reasoning: Captures and exposes Deepseek's thinking process

Installation

You can use Deepseek Thinker MCP directly with npx or install it locally.

Using npx (Recommended)

No installation needed - the server will be run on demand using npx.

Local Installation

npm install deepseek-thinker-mcp

Configuration

The server requires certain environment variables to be set depending on which mode you want to use.

OpenAI API Mode Configuration

Set these environment variables:

API_KEY=<Your OpenAI API Key>
BASE_URL=<API Base URL>

Ollama Mode Configuration

Set this environment variable to use a local Ollama installation:

USE_OLLAMA=true

Usage with Claude Desktop

OpenAI API Mode Integration

Add this configuration to your claude_desktop_config.json:

{
  "mcpServers": {
    "deepseek-thinker": {
      "command": "npx",
      "args": [
        "-y",
        "deepseek-thinker-mcp"
      ],
      "env": {
        "API_KEY": "<Your API Key>",
        "BASE_URL": "<Your Base URL>"
      }
    }
  }
}

Ollama Mode Integration

If you want to use the local Ollama server:

{
  "mcpServers": {
    "deepseek-thinker": {
      "command": "npx",
      "args": [
        "-y",
        "deepseek-thinker-mcp"
      ],
      "env": {
        "USE_OLLAMA": "true"
      }
    }
  }
}

Using a Local Installation

If you've installed the package locally:

{
  "mcpServers": {
    "deepseek-thinker": {
      "command": "node",
      "args": [
        "/your-path/deepseek-thinker-mcp/build/index.js"
      ],
      "env": {
        "API_KEY": "<Your API Key>",
        "BASE_URL": "<Your Base URL>"
      }
    }
  }
}

Available Tools

get-deepseek-thinker

This tool performs reasoning using the Deepseek model.

Input Parameters:

• originPrompt (string): The user's original prompt

Returns: A structured text response containing Deepseek's reasoning process

Troubleshooting

Request Timeout Errors

If you see an error like "MCP error -32001: Request timed out", this indicates that:

• The Deepseek API response is too slow
• The reasoning output is too long for the MCP server's timeout limit

In these cases, try simplifying your query or adjusting timeout settings if possible.

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.