← Back to Skills Marketplace
174
Downloads
0
Stars
1
Active Installs
4
Versions
Install in OpenClaw
/install wechat-fetch
Description
微信公众号文章抓取工具 v3.0 - Lite轻量版 + Playwright版，支持免登录、批量抓取、图片下载、多格式输出
Usage Guidance
This package appears to do what it claims (scrape public WeChat articles, download images, and save outputs). Before installing/running: (1) Review and be comfortable with files being written to the output directory and a browser profile stored under ~/.openclaw/browser/wechat-fetch-v3 (this may contain cookies/session info if you use Cookie mode). (2) Only scrape content you are permitted to access — scraping can violate site terms of service or copyright. (3) When using the Playwright mode, expect a Chromium download during setup. (4) Run the code in an isolated environment (virtualenv/container) if you do not fully trust the source, and inspect the scripts yourself if you have concerns about saved cookie data or filesystem paths. (5) Consider rate limits and use delays/batch sizes to avoid triggering anti-bot measures.
Capability Analysis
Type: OpenClaw Skill
Name: wechat-fetch
Version: 3.0.0
The skill bundle provides a legitimate tool for scraping WeChat articles, offering both a lightweight 'requests' version and a 'Playwright' version for handling complex pages. The code in 'scripts/wechat_fetch_lite.py' and 'scripts/wechat_fetch_v3.py' is well-structured, aligns perfectly with the stated purpose, and contains no evidence of data exfiltration, malicious execution, or prompt injection. It uses standard libraries (BeautifulSoup, Playwright, Requests) and follows common scraping patterns, including local image storage and batch processing.
Capability Assessment
Purpose & Capability
Name/description (WeChat article fetcher) match the included scripts, CLI examples, and Python API. Required dependencies (requests/BeautifulSoup and optional Playwright) make sense for the Lite and Playwright variants. Nothing in the manifest or code appears to require unrelated system access or extraneous credentials.
Instruction Scope
SKILL.md instructs the agent to fetch article URLs, optionally download images, write outputs (markdown/html/json/txt), and use Playwright for JS-rendered pages; the code follows those steps. The skill reads a provided batch file of URLs and writes output files and an images/ subdirectory. It does not contact any unexpected third-party endpoints beyond the target article/image hosts. Note: the Playwright mode uses a persistent browser context (local user_data_dir) to store cookies/profile if Cookie mode is used.
Install Mechanism
There is no packaged install spec; README/SKILL.md recommend pip install of dependencies and 'playwright install chromium' for the Playwright mode. That is a standard approach for Python/Playwright projects. The installation will download Chromium when Playwright is installed, which is expected for the Playwright variant.
Credentials
The skill does not request environment variables or other credentials. However, the Playwright variant creates a persistent browser profile under ~/.openclaw/browser/wechat-fetch-v3 to store cookies and session data when using persistent (Cookie) mode; that could retain authentication cookies and thus access private content if you log in. This persistence is proportionate to the stated Cookie-mode feature but is worth awareness.
Persistence & Privilege
always:false and the skill is user-invocable (normal). The only notable persistence is the creation of a per-skill user_data_dir in the user's home to hold a persistent browser context for Cookie mode. The skill does not modify other skills or system-wide agent configs.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install wechat-fetch - After installation, invoke the skill by name or use
/wechat-fetch - Provide required inputs per the skill's parameter spec and get structured output
Version History
v3.0.0
Major release: Adds Lite版、免登录模式、批量/图片下载、多格式输出，优化结构。
- Lite版、Playwright版双实现，适合不同环境和需求
- 支持免登录公开抓取及批量下载
- 自动文章图片下载、本地存储
- 输出格式支持 Markdown/HTML/JSON/TXT
- 删除旧代码，精简依赖，取消 Cookie 监控功能
- 提供全新批量抓取报告和命令行参数
v2.0.1
v2.0.1: 修复 SKILL.md，移除私有仓库链接
v2.0.0
WeChat-fetch 2.0.0 introduces persistent browser context and cookie auto-monitoring:
- Uses Playwright with persistent browser context to reuse logged-in cookies—no need to scan QR every fetch
- Supports headless mode and improved reliability against anti-bot protections
- Adds cookie auto-monitor/alert scripts for session maintenance
- Expanded output formats: Markdown, HTML, JSON, and TXT
- Updates docs; removes old, redundant SKILL.md
v1.0.0
Initial version release.
- Fetches WeChat public account articles and saves them in Markdown format.
- Uses Playwright and BeautifulSoup to handle dynamic content loading.
- Supports extracting article title, public account name, publish time, main content, and images.
- Offers output in Markdown, JSON, or plain text formats.
- Includes CLI and Python (sync & async) API usage.
- Provides configuration options for browser behavior and output.
Metadata
Frequently Asked Questions
What is Wechat Fetch Publish?
微信公众号文章抓取工具 v3.0 - Lite轻量版 + Playwright版，支持免登录、批量抓取、图片下载、多格式输出. It is an AI Agent Skill for Claude Code / OpenClaw, with 174 downloads so far.
How do I install Wechat Fetch Publish?
Run "/install wechat-fetch" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Wechat Fetch Publish free?
Yes, Wechat Fetch Publish is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Wechat Fetch Publish support?
Wechat Fetch Publish is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Wechat Fetch Publish?
It is built and maintained by JackyFan (@jackyfan01); the current version is v3.0.0.
More Skills