home / mcp / weixin mcp server
能够让大模型阅读微信公众号文章，使用浏览器模拟绕过反爬虫。
Configuration
View docs{
"mcpServers": {
"bwkyd-wexin-read-mcp": {
"command": "python",
"args": [
"C:/Users/你的用户名/Desktop/wx-mcp/wx-mcp-server/src/server.py"
]
}
}
}This minimal MCP lets you fetch and analyze content from WeChat public article pages by delegating the task to a model, streamlining how you extract titles, authors, publish times, and main content for summarization or analysis.
You enable the Weixin article reader MCP as a local server and then issue requests from your MCP client. Ask it to read a WeChat article by providing the article URL, and your client will receive structured article data (title, author, publish time, and content) that you can summarize, translate, or analyze with your model.
Prerequisites: Python 3.10 or newer must be installed on your system.
1) Install Python dependencies using the project’s requirements file.
2) Create the MCP server configuration to run the Python script locally.
{
"mcpServers": {
"weixin_reader": {
"command": "python",
"args": [
"C:/Users/你的用户名/Desktop/wx-mcp/wx-mcp-server/src/server.py"
]
}
}
}配置和运行细节：该 MCP 以本地 stdio 方式运行，使用 Python 作为运行时。你需要提供一个命令来启动服务器，以及（可选）环境变量来适配你的系统。下方示例展示了从本地运行的配置。
工作流程要点：当你向模型发送一个 URL 和需求时，模型会调用 read_weixin_article 工具，MCP 获取文章内容并把结果发回给模型以生成自然语言输出。
示例调用（与客户端集成时）将返回一个结构化对象，包括标题、作者、发布时间、正文等字段，帮助你快速生成摘要或分析。
工具说明：read_weixin_article(url: str) 读取微信公众号文章内容，参数是文章 URL，返回包含 title、author、publish_time、content 等字段的结构化数据。
常见注意事项：该工具仅用于个人学习与研究，遵循平台服务条款，避免高频爬取。
请在受控环境中运行本 MCP，避免在公开网络暴露敏感信息。若需要对外暴露，请实现适当的认证和访问控制。
Reads a WeChat public article given its URL and returns a structured object with the title, author, publish time, and content.