legal-text-format

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.80). The skill explicitly ingests and processes third-party web content fetched by other tools (e.g., "用户提供微信公众号链接 → AI 使用 wechat-article-fetch 抓取 → AI 调用 legal-text-format 格式化" and "读取指定路径的 Markdown 文件内容作为输入" in SKILL.md), so untrusted/public user-generated content from WeChat pages and arbitrary webpages is read and used to drive formatting decisions and filenames, enabling indirect prompt injection risk.