ai-ecommerce-expert-local-product-image-edit

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

• Third-party content exposure detected (medium risk: 0.30). IMIVA MCP 工作流在运行时通过 create_smart_refine_task 接收调用方（用户/第三方）提供的 prompt/参数并将其作为自然语言编辑指令直接送入 IMIVA MCP 客户端，因此外部作者可在提交任务时注入任意文本。

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 download and execute the external npm package "@infimind/ecom-content-cli@latest", which fetches and runs remote code required for MCP operations.