ai-ecommerce-expert-viral-video-product-adaptation

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

• Third-party content exposure detected (medium risk: 0.30). 在脚本 scripts/imiva_mcp.py 运行时，通过 invoke(method) 启动 @infimind/ecom-content-cli@latest 后把调用参数（由用户在 --args/additionalRequirements 等字段填写的免费文本）以 tools/call 的 arguments 形式发送给 IMIVA MCP 工具，进而由 IMIVA 侧 LLM 读取这些文本内容。

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

• Potentially malicious external URL detected (high risk: 0.90). The skill invokes and executes a remote npm package at runtime via npx ("@infimind/ecom-content-cli@latest"), which causes external code to be fetched and run (e.g., scripts/imiva_mcp.py starts npx with PACKAGE), so the npm package string "@infimind/ecom-content-cli@latest" is flagged.