ai-ecommerce-expert-xiaohongshu-ecommerce-content

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

• Third-party content exposure detected (medium risk: 0.30). Skill 运行时通过 scripts/imiva_mcp.py 的工具调用把用户在 --args（如 productName、sellingPoints、targetAudience、images 等）中提交的自由文本/素材路径传入 IMIVA MCP，并非从浏览器/队列/外部内容“先选定项再取回”。

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

• Potentially malicious external URL detected (high risk: 0.90). The skill invokes npx at runtime to fetch and execute the npm package "@infimind/ecom-content-cli@latest" (see scripts/imiva_mcp.py), so remote code is fetched and run as a required dependency.