Zh MCP Server

🚀 zh_mcp_server Introduction

This is a Model Context Protocol (MCP) server for publishing articles on Zhihu. Users can collaborate with large models through this service to generate articles and automatically publish them on Zhihu.

🚀 Quick Start

📦 Installation

1. Clone the code

git clone https://github.com/Victorzwx/zh_mcp_server.git

2. Environment configuration prerequisites

It is recommended to run in a Windows environment. The Python version requirement is >= 3.10.

• Method 1: Configure according to the dependencies in the requirements.txt file:
  • selenium>=4.0.0
  • requests>=2.25.1
  • mcp>=0.1.0
  • webdriver-manager>=3.8.0

You can install them using the following command:

pip install -r requirements.txt

Then install ChromeDriver. This project depends on Google Chrome with a version number of 134.0.6998.166. You need to manually query and download the ChromeDriver version that matches the Google Chrome on your current system.

npx @puppeteer/browsers install chromedriver@134.0.6998.166

• Method 2: Run the setup_environment.py script. If this method fails, it may be due to an incorrect ChromeDriver version. It is recommended to reinstall using Method 1.

python setup_environment.py

3. Save personal cookies

Run the following command in the code folder to save cookies:

python -m zh_mcp_server.__login__

• Note: After running, Google Chrome will automatically open. Please log in using your mobile phone number and click to get the verification code.
• It's very important to enter the verification code you received in the Terminal!

4. Configure the MCP service in the MCP client (e.g., Cherry Studio)

Run it in Python:

"zh_mcp_server": {
      "command": "python",
      "args": [
        "-m",
        "zh_mcp_server"
      ]
 }

Then you can use the service.

If you call the MCP service through other languages (e.g., Java code based on Spring AI), pay attention to the encoding method to avoid garbled characters:

"zh_mcp_server": {
      "command": "D:\\aconda\\python.exe",
      "args": [
        "-m",
        "zh_mcp_server",
        "--encoding=utf-8"
      ],
      "env": {
        "PYTHONIOENCODING": "utf-8"
      }
    }

💻 Usage Examples

Basic Usage

The above steps show the basic process of using the zh_mcp_server.

Advanced Usage

Debugging

If you need to debug the specific process of the large model calling the MCP service or visualize browser operations, turn off the headless browser mode:

poster = ZhuHuPoster(path, headless=True)##Set to False if you need to debug

The code is located in server.py.

📚 Additional Information

CSDN

Personal CSDN blog

WeChat