tylersuard/mcp-123

MCP-123

运行MCP服务器和客户端（各2行）的最简单方法

---

🚀 特色

• 超最小配置：在两行内启动服务器或客户端。

• 简单的工具创建：在你的tools.py文件中写入普通函数——不需要装饰工具或特殊封装器——它们会作为工具被包含在你的MCP服务器上，自动使用。

• OpenAI 集成：客户端使用您的 OpenAI API 密钥回答问题，并根据需要调用工具。

---

🖥️ 快速开始

1.安装要求

2.创建你的工具

在tools.py中定义你的函数。不需要装饰器，它们会自动作为工具添加到你的MCP服务器。例如：

3.启动MCP服务器（2行）

4.设置MCP客户端（2行）

5.使用 MCP 客户端

6.完成后关闭MCP客户端

---

🚀 特色

• 超最小配置：在两行内启动服务器或客户端。

• 简单工具创建：在你的tools.py文件中写入普通函数——无需装饰器或特殊封装器——这些函数会作为工具被纳入MCP服务器自动使用。

MCP-123

The easiest way to run an MCP server & client (2 lines each)

---

🚀 Features

• Ultra-minimal setup: Start a server or client in 2 lines.

• Easy tool creation: Write normal functions in your tools.py file—no decorators or special wrappers needed—and they get included as tools that your MCP server can use automatically.

• OpenAI integration: The client uses your OpenAI API key to answer questions, calling tools as needed.

---

🖥️ Quickstart

1. Install Requirements

pip install -r requirements.txt

2. Create Your Tools

Define your functions in tools.py. No decorators needed, they are automatically added to your MCP server as tools. For example:

def add(a: int, b: int) -> int:

    """Add two numbers."""

    return a + b

3. Start the MCP Server (2 lines)

from mcp123 import server

server.run_server("tools.py", port=9999)

4. Set up the MCP Client (2 lines)

from mcp123.client import McpClient

client = McpClient("http://localhost:9999", "sk-...your OpenAI key...")

5. Use the MCP Client

answer = client.ask("Add 15 and 14.")

print("Answer:", answer)

6. Close the MCP Client when you are done

client.close()

---

🚀 Features

• Ultra-minimal setup: Start a server or client in 2 lines.

• Easy tool creation: Write normal functions in your tools.py file—no decorators or special wrappers needed—and they get included as tools that your MCP server can use automatically.

• OpenAI integration: The client uses your OpenAI API key to answer questions, calling tools as needed.

---

📝 How It Works

• Server: Loads all top-level functions from tools.py and exposes them as MCP tools via HTTP.

• Client: Discovers available tools, sends prompts to OpenAI, and automatically calls tools if needed.

---

🛠️ Example Output

When you run the client, you’ll see:

Tools discovered:

 [ ...list of tools... ]

Answer: 29

---

🔑 Requirements

• Python 3.11+

• OpenAI API key (for the client)

---

📢 Why MCP123?

• Zero boilerplate: No need to write schemas or wrappers—just write functions.

• LLM-native: Designed for seamless LLM tool use.

• Extensible: Add more tools by simply adding functions.

---

🤝 Credits

• Built with FastMCP

• Inspired by the Model Context Protocol (MCP)

---

📬 Feedback & Contributions

Pull requests and issues are welcome, but only if they are in ALL-CAPS.