ai-ecommerce-expert-shoes-bags-ecommerce-content

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

• Third-party content exposure detected (medium risk: 0.30). 在 scripts/imiva_mcp.py 的 invoke() 运行时，LLM 所生成的 user 提交 JSON 参数会被直接送入 IMIVA MCP 工具（method: tools/call，params: {name, arguments}），因此外部用户自由文本（如 coreSellingPoints/usageScenario 等）会在未先选择特定条目的情况下被摄入并用于生成任务。

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

• Potentially malicious external URL detected (high risk: 1.00). The skill runs npx to fetch and execute the npm package "@infimind/ecom-content-cli@latest" at runtime (see PACKAGE and subprocess invocation), which downloads and runs remote code that the skill depends on.