验证码识别 MCP 服务

简介

本项目基于 ddddocr 和 Model Context Protocol (MCP) Python SDK，实现了一个通过 MCP 协议对外提供验证码图片识别服务的智能后端。

依赖

• ddddocr
• mcp[fastmcp]

安装

1. 克隆本项目

   git clone <your-repo-url>
   cd <your-project-dir>
   

2. 安装依赖

   pip install ddddocr mcp[fastmcp]
   

使用方法

启动服务端

python main.py

客户端调用示例

from mcp import ClientSession, StdioServerParameters
from mcp.client.stdio import stdio_client
import asyncio

async def main():
    server_params = StdioServerParameters(
        command="python",
        args=["main.py"],
        env=None,
    )
    async with stdio_client(server_params) as (read, write):
        async with ClientSession(read, write) as session:
            await session.initialize()
            result = await session.call_tool(
                "recognize_captcha",
                arguments={"image_path": "E:/code/bug/service/ocr/captchaImage.jpg"}
            )
            print("识别结果:", result)

if __name__ == "__main__":
    asyncio.run(main())

工具接口

• recognize_captcha(image_path: str)
  识别本地图片验证码，返回内容对象

客户端使用

{
    "mcpServers": {
        "ocr": {
            "command": "F:\\AI\\anaconda3\\Scripts\\uv.exe",
            "args": [
                "run",
                "--with",
                "mcp",
                "mcp",
                "run",
                "E:\\code\\bug\\service\\ocr\\mcp-server-demo\\main.py"
            ]
        }
    }
}

License

MIT