How do I install the Wx2md skill?

Run npx skillmds@latest add zhouyinlong-lab/wx2md in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Wx2md safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Wx2md?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Wx2md free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

zhouyinlong-lab (@zhouyinlong-lab) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Wx2md by zhouyinlong-lab — Agent Skill

Name: Wx2md
Author: zhouyinlong-lab

Wx2md

Export WeChat Official Account article links from mp.weixin.qq.com to local Markdown with front matter and localized images. Use when the user wants a CLI/script or workflow for saving WeChat public account articles as Markdown for personal learning, archive, AI input, Obsidian notes, or knowledge management. Supports Playwright-based page loading, WeChat DOM extraction, image downloading, Markdown conversion, batch URL files, and validation of export output.

File contents

WX2MD

Use this skill to fetch a WeChat Official Account article and export it as a local Markdown note with metadata and images.

Tool

The bundled CLI lives in this skill folder:

cd wx2md
pip install -r requirements.txt
playwright install chromium
pip install -e .
wx2md "https://mp.weixin.qq.com/s/xxxx"

Direct script mode also works:

cd wx2md
python src/main.py "https://mp.weixin.qq.com/s/xxxx" --output ./output/articles

Workflow

Validate the input is either a WeChat article URL or a text file containing URLs.
Use Playwright Chromium to load the page because WeChat pages often need browser DOM execution.
Extract metadata with WeChat selectors:
- #activity-name for title
- #js_name for account/author
- #publish_time for publish time
- #js_content for article body
Normalize image URLs using data-src, then src, then data-backsrc.
Download images with User-Agent and Referer headers unless --no-images is set.
Replace image URLs with local relative paths.
Convert the article body HTML to Markdown.
Write index.md under output/articles/<sanitized title>/.

Commands

wx2md "https://mp.weixin.qq.com/s/xxxx"
wx2md "https://mp.weixin.qq.com/s/xxxx" --output ./output/articles
wx2md "https://mp.weixin.qq.com/s/xxxx" --no-images
wx2md "https://mp.weixin.qq.com/s/xxxx" --format obsidian
wx2md "https://mp.weixin.qq.com/s/xxxx" --title "custom title"
wx2md urls.txt --output ./output/articles

Output

Default output:

output/
  articles/
    Article Title/
      index.md
      images/
        001.jpg
        002.png

Obsidian output uses attachments/ for images and adds simple front matter tags.

Verification

After export, check:

index.md exists.
Front matter includes title, author, source, URL, created, and exported_at.
Local image files exist unless --no-images was used.
Markdown image links are relative and render from index.md.
The body contains the article's major headings, paragraphs, quotes, and images.

Limits

This tool does not bypass login, CAPTCHA, region restrictions, deleted content, paywalls, or anti-abuse controls. If the browser cannot load the article or #js_content is empty, report a clear error and preserve the original link for manual handling.

Use only for personal learning, backup, and knowledge management. Preserve original source and URL.

---name: wx2mddescription: Export WeChat Official Account article links from mp.weixin.qq.com to local Markdown with front matter and localized images. Use when the user wants a CLI/script or workflow for saving WeChat public account articles as Markdown for personal learning, archive, AI input, Obsidian notes, or knowledge management. Supports Playwright-based page loading, WeChat DOM extraction, image downloading, Markdown conversion, batch URL files, and validation of export output.---# WX2MDUse this skill to fetch a WeChat Official Account article and export it as a local Markdown note with metadata and images.## ToolThe bundled CLI lives in this skill folder:```powershellcd wx2mdpip install -r requirements.txtplaywright install chromiumpip install -e .wx2md "https://mp.weixin.qq.com/s/xxxx"```Direct script mode also works:```powershellcd wx2mdpython src/main.py "https://mp.weixin.qq.com/s/xxxx" --output ./output/articles```## Workflow1. Validate the input is either a WeChat article URL or a text file containing URLs.2. Use Playwright Chromium to load the page because WeChat pages often need browser DOM execution.3. Extract metadata with WeChat selectors: - `#activity-name` for title - `#js_name` for account/author - `#publish_time` for publish time - `#js_content` for article body4. Normalize image URLs using `data-src`, then `src`, then `data-backsrc`.5. Download images with `User-Agent` and `Referer` headers unless `--no-images` is set.6. Replace image URLs with local relative paths.7. Convert the article body HTML to Markdown.8. Write `index.md` under `output/articles/<sanitized title>/`.## Commands```powershellwx2md "https://mp.weixin.qq.com/s/xxxx"wx2md "https://mp.weixin.qq.com/s/xxxx" --output ./output/articleswx2md "https://mp.weixin.qq.com/s/xxxx" --no-imageswx2md "https://mp.weixin.qq.com/s/xxxx" --format obsidianwx2md "https://mp.weixin.qq.com/s/xxxx" --title "custom title"wx2md urls.txt --output ./output/articles```## OutputDefault output:```textoutput/ articles/ Article Title/ index.md images/ 001.jpg 002.png```Obsidian output uses `attachments/` for images and adds simple front matter tags.## VerificationAfter export, check:- `index.md` exists.- Front matter includes title, author, source, URL, created, and exported_at.- Local image files exist unless `--no-images` was used.- Markdown image links are relative and render from `index.md`.- The body contains the article's major headings, paragraphs, quotes, and images.## LimitsThis tool does not bypass login, CAPTCHA, region restrictions, deleted content, paywalls, or anti-abuse controls. If the browser cannot load the article or `#js_content` is empty, report a clear error and preserve the original link for manual handling.Use only for personal learning, backup, and knowledge management. Preserve original source and URL.

Frequently asked questions