deepseek-claude

DeepSeek-Claude MCP Server extends Claude Desktop by wiring in DeepSeek R1's reasoning engine to handle complex multi-step reasoning tasks. Once configured, Claude can send queries that require advanced deduction to the MCP server, which processes them with the DeepSeek engine and returns structured reasoning results that Claude can incorporate into its final answer. The workflow relies on the MCP client (UV in this setup) to start and manage the server, and Claude Desktop to detect and invoke the MCP server as a tool during conversations. The server is designed to provide precise, reasoned outputs by leveraging DeepSeek R1 for depthful thinking, while Claude handles natural language generation and integration of the reasoning into cohesive responses.

Prerequisites

• Python 3.12 or higher
• uv package manager available on your system
• DeepSeek API key (sign up at the DeepSeek Platform)

Local installation steps

1. Clone the repository git clone https://github.com/harshj23/deepseek-claude-MCP-server.git cd deepseek-claude-MCP-server

2. Ensure UV is set up on your platform

   • Windows (PowerShell): powershell -ExecutionPolicy ByPass -c "irm https://astral.sh/uv/install.ps1 | iex"
   • Mac: curl -LsSf https://astral.sh/uv/install.sh | sh

3. Create and activate a Python virtual environment uv venv source .venv/bin/activate

4. Install required dependencies uv add "mcp[cli]" httpx

5. Set up your DeepSeek API key Obtain your API key from here: https://platform.deepseek.com/api_keys (Set this in your environment as needed by the server or configuration.)

6. Configure the MCP server Create or edit claude_desktop_config.json (as shown in the reference) to include:

   { "mcpServers": { "deepseek-claude": { "command": "uv", "args": [ "--directory", "C:\ABSOLUTE\PATH\TO\PARENT\FOLDER\deepseek-claude", "run", "server.py" ] } } }

7. Run the server uv run server.py

8. Test setup

   • Restart Claude Desktop and ensure the tools icon for the MCP server is visible
   • If the server isn’t visible, consult the troubleshooting guide provided in the project or documentation

Alternative: Smithery installation (if you prefer the automated path)

• npx -y @smithery/cli install @HarshJ23/deepseek-claude-MCP-server --client claude

Prerequisites summary: Python 3.12+, uv, DeepSeek API key, network access to DeepSeek, and a configured claude_desktop_config.json pointing to the local run command.

Tips and notes:

• Keep your DeepSeek API key secure. Do not commit it to public repos.
• Ensure the path in the configuration (--directory) points to the directory containing server.py for the DeepSeek-Claude MCP server.
• If Claude Desktop cannot detect the MCP server, verify that uv is properly installed and that the server process is running (check logs for errors).
• The DeepSeek R1 integration handles complex reasoning; you may see outputs wrapped or annotated by DeepSeek-specific markers depending on the version—treat these as structured reasoning to be incorporated by Claude.
• You can customize environment variables as needed by your deployment (e.g., API keys, endpoints). If using a container or remote host, ensure network access to the DeepSeek API key endpoint is available.
• For troubleshooting, refer to the project’s quickstart or troubleshooting guide linked in the README and ensure the correct Python environment is active when running uv run server.py.