Wechat Art Parser

This MCP server parses WeChat public account articles, extracting the title, author, publish time, and plain-text content. It focuses on text-based analysis and supports deep semantic analysis via a built-in LLM integration. It runs using the Standard I/O MCP transport and can be composed with a client like Cursor to provide automated article understanding workflows.

How to use

You run the server locally and connect it to an MCP client to parse WeChat public account articles. The server exposes a stdio-based MCP endpoint, so you launch it and configure a client to start a managed parsing session. Once connected, you can parse an article URL to obtain its title, author, publish time, a clean text body, and a detailed summary generated by an LLM. You can also request deep semantic analysis to identify core viewpoints, article structure, and writing style.

How to install

Prerequisites: you need Python and a virtual environment tool. Install dependencies from the provided requirements file, then run the server script.

# Create a virtual environment
python -m venv venv

# Activate the virtual environment
# macOS/Linux:
source venv/bin/activate
# Windows:
venv\Scripts\activate

pip install -r requirements.txt

If you are behind a slow network, use a mirror:

pip install -r requirements.txt -i https://pypi.tuna.tsinghua.edu.cn/simple

# Start the MCP server
python server.py

Configuration and running with Cursor

Configure Cursor to connect to the MCP server using stdio transport. Provide the absolute path to the server script and specify the Python command.

{
  "mcpServers": {
    "wechat_art_parser": {
      "command": "python3",
      "args": ["/绝对路径/wechat-article-parser-mcp/server.py"],
      "env": {
        "ZHIPU_API_KEY": "YOUR_API_KEY"
      }
    }
  }
}

Notes on LLM integration and API keys

If you want deep semantic analysis, you can enable an external LLM service by providing an API key. The API key should be supplied through the environment variable ZHIPU_API_KEY in your MCP config. This enables analyze_with_llm operations that produce structured, multi-faceted analysis of the article.

Available tools

parse_article

Parses a WeChat article URL to extract title, author, publish_time, and a plain-text summary including a detailed, long-form summary generated via LLM.

analyze_with_llm

Performs deep semantic analysis and viewpoint/structure extraction using a large language model. Supports multiple analysis types and models, and can save results to a Markdown file.