DeepSeek MCP Server
Connects MCP clients to DeepSeek API, including DeepSeek-R1 reasoning with visible chain-of-thought. Provides four tools: generate, chat, reason, and list models.
README
DeepSeek MCP Server
A small, clean Model Context Protocol server that connects any MCP client — Claude Desktop, Cursor, Cline, Cowork — to the DeepSeek API, including DeepSeek-R1 reasoning with a visible chain-of-thought.
<p align="center"><img src="demo.gif" alt="deepseek-mcp-server demo" width="760"></p>
One Python file, four tools, no framework lock-in. Bring your own DeepSeek key and you get fast V3 generation, multi-turn chat, and R1 deep reasoning that shows its work — straight from your assistant.
✨ Tools
| Tool | What it does |
|---|---|
deepseek_generate |
Single-turn text with DeepSeek-V3 (system instruction, temperature, JSON mode) |
deepseek_chat |
Multi-turn conversation; switch to deepseek-reasoner for step-by-step thinking |
deepseek_reason |
Hard problems with DeepSeek-R1 — returns the reasoning trace and the answer |
deepseek_list_models |
List models available to your key |
🚀 Quick start
Option A — one-click (Windows, Claude Desktop)
# in the repo folder, right-click install.ps1 -> "Run with PowerShell"
./install.ps1
It asks for your API key, installs dependencies, and wires up Claude Desktop. Restart Claude and you're done.
Option B — manual (any OS / any MCP client)
git clone https://github.com/deepzhun/deepseek-mcp-server.git
cd deepseek-mcp-server
pip install -r requirements.txt
export DEEPSEEK_API_KEY="your-api-key" # Windows: setx DEEPSEEK_API_KEY "your-api-key"
python deepseek_mcp.py
Get an API key at https://platform.deepseek.com/api_keys. DeepSeek's endpoint is OpenAI-compatible.
🔌 Add to your MCP client
Copy config.example.json and drop the server block into your client config,
replacing the placeholder key:
{
"mcpServers": {
"deepseek": {
"command": "python",
"args": ["/absolute/path/to/deepseek_mcp.py"],
"env": { "DEEPSEEK_API_KEY": "YOUR_DEEPSEEK_API_KEY_HERE" }
}
}
}
💬 Usage examples
- "Use deepseek to draft a product update in Chinese."
- "Ask deepseek_reason to solve this logic puzzle and show its reasoning."
- "Use deepseek_chat to continue our debugging conversation."
🔐 Security
- The key is read from the
DEEPSEEK_API_KEYenvironment variable — never hard-code it. config.jsonand.envare git-ignored so a real key can't be committed by accident.- Rotate your key in the DeepSeek console if it is ever exposed.
🛠️ Development
pip install -r requirements.txt
python -c "import deepseek_mcp; print(deepseek_mcp.mcp.name)" # smoke test
Single-file server (deepseek_mcp.py): typed Pydantic inputs, consistent error
handling, MCP tool annotations. A sibling of gemini-mcp-server — same clean shape, different provider.
📄 License
MIT © deepzhun (深准). Contributions welcome — open an issue or PR.
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。