Read WeChat Official Account articles and convert to Markdown — no login required.
OpenClaw Skill — works with OpenClaw AI agents
Takes any mp.weixin.qq.com article URL and extracts the full content as structured Markdown, including title, author, publish date, and body text. Uses httpx with mobile WeChat User-Agent and upstream proxy to fetch article HTML (server-side rendered, no browser needed), then converts to Markdown with markdownify.
# Install via ClawHub (recommended)
openclaw skill install wechat-reader
# Or clone this repo
git clone https://github.com/rrrrrredy/wechat-reader.git ~/.openclaw/skills/wechat-reader
# Install dependencies
bash scripts/setup.sh- Full article extraction: Title, author, publish date, and body text
- Markdown output: Clean, structured Markdown with proper headings and formatting
- No browser needed: Uses
httpxHTTP client directly — WeChat articles are server-side rendered, no JavaScript execution required - Proxy support: Built-in upstream proxy configuration for network access
- No login required: Works with public articles without authentication
- Error resilience: Timeout handling and fallback selectors
- Image limitation: Images are output as empty
![]()placeholders — WeChat CDN uses anti-hotlinking, image URLs cannot be directly downloaded
触发词：
读这篇公众号、公众号文章、微信文章微信公众号链接、mp.weixin.qq.com读取公众号内容、公众号全文
Example:
"帮我读一下这篇公众号文章：https://mp.weixin.qq.com/s/xxxxx"
不适用： 需要登录/付费/限阅的内容；非微信公众号链接（用 web-access 处理其他网页）。
wechat-reader/
├── SKILL.md # Main skill definition
├── scripts/
│ ├── setup.sh # Dependency installer
│ └── fetch_article.py # Article fetcher & converter
├── references/
│ └── technical-notes.md # Implementation notes
└── .gitignore
- OpenClaw agent runtime
- Python 3.8+
httpx(HTTP client with proxy support)markdownify(HTML to Markdown)- HTTP upstream proxy for WeChat server access (direct IP blocked)