← Back to Skills Marketplace
264
Downloads
1
Stars
0
Active Installs
4
Versions
Install in OpenClaw
/install wechat-article-fetcher-safe
Description
使用 Puppeteer + Chrome 无头模式抓取微信公众号文章，支持移动端伪装，提取全文及元信息并自动保存文本。
Usage Guidance
This skill appears to do what it claims: locally launch Chrome via puppeteer-core, render a WeChat article, extract content, and save files. Before installing, note: (1) you must run npm install (puppeteer-core) and have a compatible local Chrome installed — puppeteer-core does not ship Chromium; (2) the tool will connect to the article URL and download images from whichever hosts the page references (page scripts may also trigger additional network activity), so the SKILL.md phrase 'no data sent to external servers' only means the skill doesn't exfiltrate additional private data deliberately; it still fetches remote page resources as part of scraping; (3) run this in a normal user account (avoid elevated privileges) and review the code if you plan to run it on sensitive systems; (4) if you want stricter network controls, run it in an isolated environment or sandbox to limit outgoing connections. Overall the package is internally consistent and not suspicious.
Capability Analysis
Type: OpenClaw Skill
Name: wechat-article-fetcher-safe
Version: 2.0.1
The skill bundle is a legitimate tool for fetching and converting WeChat articles to Markdown/HTML using Puppeteer. The core logic in `fetch-wechat-article.js` performs standard web scraping, DOM traversal, and image downloading to the local file system, consistent with the stated purpose in `SKILL.md`. No evidence of data exfiltration, unauthorized network calls, or malicious instructions was found.
Capability Assessment
Purpose & Capability
Name/description (WeChat article fetcher) match the code and package.json: it uses puppeteer-core + local Chrome, opens the article URL, extracts text/metadata, and optionally downloads images. Declared capabilities (file_read, file_write, exec, browser) are appropriate for the task.
Instruction Scope
SKILL.md instructions stick to launching the script and running Puppeteer. One important caveat: the README/SKILL.md claim '不发送数据到任何外部服务器' is somewhat misleading — the skill necessarily connects to the provided mp.weixin.qq.com URL and to remote image hosts, and page JavaScript executed in the loaded page can make additional network requests. The instructions do not ask the agent to read unrelated files or environment variables.
Install Mechanism
There is no formal install spec in the skill metadata, but package.json declares puppeteer-core. Installation requires running npm install locally (as documented). puppeteer-core does not bundle Chromium, so a local Chrome is required (the script auto-detects common paths). No downloads from unknown personal servers or URL shorteners were observed in the code.
Credentials
The skill requests no environment variables or credentials. It only needs access to a local Chrome binary and filesystem write access to save outputs — both are proportionate to its purpose.
Persistence & Privilege
always: false and normal agent invocation. The skill does not request permanent/global presence or attempt to modify other skills. File write and exec permissions are expected for saving results and launching Chrome.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install wechat-article-fetcher-safe - After installation, invoke the skill by name or use
/wechat-article-fetcher-safe - Provide required inputs per the skill's parameter spec and get structured output
Version History
v2.0.1
- Removed test files and legacy test outputs for a cleaner repository.
- Added a new GitHub-focused README and project metadata file.
- Updated documentation to reflect new repo location and privacy guarantees.
- Improved clarity: streamlined feature description, file structure, and troubleshooting sections.
- Removed unused OCR-related dependencies and instructions.
v2.0.0
**Major update: project files and auxiliary docs removed for a cleaner, production-ready release.**
- Removed four files: .clawhub/origin.json, .gitignore, README.github.md, UPLOAD.md
- Streamlined repository structure by deleting project meta/config and upload guideline files
- Skill documentation (SKILL.md) remains unchanged
- Core functionality and user-facing documentation are unaffected
v1.1.0
**WeChat Article Fetcher Safe v1.1.0**
- Renamed skill to "wechat-article-fetcher-safe" to highlight improved privacy protection.
- Added new test output files and a unified test runner (run-tests.js) for streamlined testing.
- Enhanced article content cleaning and failure detection mechanisms.
- Implemented automatic Chrome path detection for better portability.
- Updated documentation, usage instructions, and file structure examples.
v1.0.0
WeChat Article Fetcher v1.0.0
- Initial release.
- Supports full-text fetching of WeChat official account articles, including metadata extraction (title, author, time).
- Uses Puppeteer with Chrome in headless mode and mobile User-Agent spoofing to bypass anti-scraping.
- Handles dynamic rendering, automatic saving to text files, error handling, and timeout protection.
- Achieved 100% success rate in 3/3 tests.
Metadata
Frequently Asked Questions
What is 微信文章抓取器 - 安全版?
使用 Puppeteer + Chrome 无头模式抓取微信公众号文章，支持移动端伪装，提取全文及元信息并自动保存文本。 It is an AI Agent Skill for Claude Code / OpenClaw, with 264 downloads so far.
How do I install 微信文章抓取器 - 安全版?
Run "/install wechat-article-fetcher-safe" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is 微信文章抓取器 - 安全版 free?
Yes, 微信文章抓取器 - 安全版 is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does 微信文章抓取器 - 安全版 support?
微信文章抓取器 - 安全版 is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created 微信文章抓取器 - 安全版?
It is built and maintained by Adgai115 (@adgai115); the current version is v2.0.1.
More Skills