Deepseek MCP Server

🚀 🧠 DeepSeek MCP Server

Enhance Claude's reasoning ability by integrating the advanced inference engine of DeepSeek R1.

🚀 Quick Start

This server enables Claude to handle complex reasoning tasks by leveraging the inference capabilities of the DeepSeek R1 model.

✨ Features

Advanced Reasoning Capabilities

• Support complex multi-step reasoning tasks.
• Designed to generate well-thought-out responses in a precise and efficient manner.
• Use the API of Infini-ai (Access Address)

Division of Labor between Components

• DeepSeek R1 (Brain) as an Advanced Reasoning Planner:
  • Plan multi-step logical analysis strategies.
  • Build cognitive frameworks.
  • Evaluate confidence and uncertainty.
  • Monitor reasoning quality.
  • Detect boundary cases and deviations.
• Claude (Executor) to Implement Reasoning Plans:
  • Execute structured analysis.
  • Implement planned strategies.
  • Provide final responses.
  • Handle user interactions.
  • Manage system integration.

📦 Installation

Prerequisites

• Python 3.12 or higher
• uv package manager
• INFINI_API_KEY For DeepSeek (Register and obtain it at Infini-ai)

1. Clone the repository

   git clone https://github.com/moyu6027/deepseek-MCP-server.git
   cd deepseek-MCP-server
   

2. Ensure UV is set up

   • Windows: Run the following command in PowerShell:

     powershell -ExecutionPolicy ByPass -c "irm https://astral.sh/uv/install.ps1 | iex"
     

   • Mac: Run the following command:

     curl -LsSf https://astral.sh/uv/install.sh | sh
     

3. Create a virtual environment

   uv venv
   source .venv/bin/activate
   

4. Install dependencies

   uv add "mcp[cli]" httpx
   

5. Set up the API key

   echo "INFINI_API_KEY=your_key_here" > .env
   

6. Install the server

   mcp install server.py -f .env
   

7. Configure the MCP server Edit the claude_desktop_config.json file and add the following configuration:

   {
       "mcpServers": {
           "deepseek-mcp": {
               "command": "uv",
               "args": [
                   "--directory",
                   "PATH_TO_DEEPSEEK_MCP_SERVER",
                   "run",
                   "server.py"
               ]
           }
       }
   }
   

8. Run the server

   uv run server.py
   

💻 Usage Examples

Starting the Server

The server will start automatically when used with Claude Desktop. Ensure that Claude Desktop is configured to detect the MCP server.

Example Workflow

1. Claude receives a question that requires advanced reasoning.
2. The question is forwarded to DeepSeek R1 for processing.
3. DeepSeek R1 returns structured reasoning content wrapped in <ant_thinking> tags.
4. Claude integrates its reasoning into the final response.

📄 License

This project is licensed under the MIT License. For more details, please refer to the LICENSE file.