ai-ecommerce-expert-sku-variant-main-image

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

• Third-party content exposure detected (medium risk: 0.30). 源自用户/调用方在运行时通过 Skill/脚本提供的自由文本参数（如 coreSellingPoints、generationStyle、usageScenario、positionTypes/positionCopyModes 等）会被组装进对 create_product_main_image_task 的 MCP 工具调用，从而成为被该工具/后端读取的输入；该流程未先选择固定条目后再摄入，而是直接把外部文本作为任务参数传入。

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

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