微信公众号阅读器

提取微信公众号文章内容，支持任意公众号文章链接的内容抓取和结构化输出。

这项技能在用途与实现上是一致的——它使用 Playwright 渲染页面并抓取文章内容。注意几点： - 安装会下载 Playwright 的浏览器二进制（需要网络与磁盘空间）； - 存在一个“stealth”脚本，专门用于绕过反爬（可能违反目标站点的服务条款或触发法律/使用政策风险）； - 技能声称有 SSRF 防护并实现了基于主机名和解析地址的检查，但没有绝对保证（不要把内部/私有网络地址或机密域名传入）； - 脚本会在 /tmp 和 ~/.openclaw/workspace 等位置写入截图/HTML；如果你在意隐私或敏感环境，建议先在隔离的沙箱/测试环境中运行或彻底审核代码。

Type: OpenClaw Skill Name: weixin-reader Version: 1.1.0 The bundle provides article extraction tools using Playwright. While the primary script (extract.py) implements robust SSRF protection via DNS resolution validation, this critical security control is missing from several other functional scripts in the bundle, including extract_generic.py, extract_notion.py, and search_article.py. This inconsistency creates a significant SSRF vulnerability, allowing the agent to be potentially manipulated into accessing internal network resources. Although there is no evidence of intentional malice or data exfiltration, the high-risk nature of unvalidated browser automation justifies a suspicious classification.