mcp -deepseek
MCP Server for deepseek integration in Claude desktop
claude mcp add --transport stdio vincentf305-mcp-server-deepseek docker run -i --rm -e DEEPSEEK_API_KEY mcp_server_deepseek \ --env DEEPSEEK_API_KEY="your_api_key_here"
How to use
This MCP server enables Claude Desktop to interact with Deepseek models running inside Docker. By deploying the server in a Docker container, Claude can send MCP requests to the 8765 port exposed by the container and obtain model inferences from the Deepseek-backed service. The Deepseek integration requires a valid API key, provided via the DEEPSEEK_API_KEY environment variable, to authorize requests to the Deepseek service. You can run the server either locally using Python or as a Docker container, and then point Claude Desktop to the configured MCP server.
To use the server from Claude Desktop, configure an MCP server entry that launches the container (or runs the local Python server) and ensure the DEEPSEEK_API_KEY is supplied. The Docker workflow builds and runs an image named mcp_server_deepseek and exposes port 8765, so Claude Desktop can reach the MCP endpoint at localhost:8765 by default. If you prefer to run the server locally, you can start it with python -m mcp_server_deepseek.server and then configure Claude Desktop to connect to http://localhost:8765 (or another host/port if you mapped differently).
How to install
Prerequisites:
- Docker installed and running
- Python 3.11 or later (for local development)
- A Deepseek API key
- Claude Desktop (for integration)
Installation steps:
-
Clone the repository: git clone https://github.com/vincentf305/mcp-server-deepseek.git cd mcp-server-deepseek
-
Install Python dependencies (for local run): python -m pip install --upgrade pip pip install -r requirements.txt
-
Set up your Deepseek API key:
- Create a .env file in the project root (optional for local run) and add: DEEPSEEK_API_KEY=your_api_key_here
- Or supply the key via the environment variable when running Docker: -e DEEPSEEK_API_KEY=your_api_key_here
-
Build and run with Docker (preferred for deployment): docker build -t mcp_server_deepseek . docker run -d
--name mcp-server-deepseek
-p 8765:8765
-e DEEPSEEK_API_KEY=your_api_key_here
mcp_server_deepseek -
Run locally (alternative): python -m mcp_server_deepseek.server
-
Validate the service:
- Ensure the container is running and listening on port 8765 ( Docker: docker ps )
- Point Claude Desktop to the MCP server endpoint (see usage notes ).
Additional notes
Tips and considerations:
- Ensure your DEEPSEEK_API_KEY is valid and has the necessary permissions for the Deepseek services you plan to use.
- When using Docker, remember to map port 8765 (or whichever port you configure) so Claude Desktop can reach the MCP endpoint.
- If you need to run multiple MCP servers, assign distinct container names and ports to avoid conflicts.
- If you update dependencies, re-build the Docker image or re-install local requirements as needed.
- Check logs for troubleshooting: Docker logs for the container, or run the local server with verbose logging if available.
- Environment variables can be extended if the MCP server supports additional config keys in the future.
Related MCP Servers
mcp-vegalite
MCP server from isaacwasserman/mcp-vegalite-server
github-chat
A Model Context Protocol (MCP) for analyzing and querying GitHub repositories using the GitHub Chat API.
nautex
MCP server for guiding Coding Agents via end-to-end requirements to implementation plan pipeline
pagerduty
PagerDuty's official local MCP (Model Context Protocol) server which provides tools to interact with your PagerDuty account directly from your MCP-enabled client.
futu-stock
mcp server for futuniuniu stock
mcp -boilerplate
Boilerplate using one of the 'better' ways to build MCP Servers. Written using FastMCP