ai-amazing-tech

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

• Third-party content exposure detected (medium risk: 0.30). Skill.md 的必需运行流程会调用 mcp_web-search__web_search 与 mcp_patent-search__patsnap_search 来获取新闻/专利/论文/政策内容，并在“技术情报简报”模块中把检索到的外部文本用于生成 HTML 报告，因此在用户输入触发后可通过外部页面/条目作者提交的自由文本产生间接提示注入风险。

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.70). The skill requires embedding and loading remote JavaScript/CSS from CDNs (https://cdn.tailwindcss.com, https://cdn.jsdelivr.net/npm/echarts@5.4.3/dist/echarts.min.js, https://fonts.googleapis.com/css2?family=Inter:wght@300;400;500;600&display=swap) as a mandatory per-report dependency, which results in executing external code at runtime in the generated reports.