ai-ecommerce-expert-seedance-2-ecommerce-video

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

• Third-party content exposure detected (medium risk: 0.30). The runtime workflow invokes the IMIVA MCP tool create_video_generation_task with a user-supplied prompt/parameters (including free-form text) via scripts/imiva_mcp.py call ... --args, meaning outsider-authored text is directly ingested by the model without first selecting a specific trusted item.

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

• Potentially malicious external URL detected (high risk: 0.90). I flagged the npm package reference "@infimind/ecom-content-cli@latest" because the skill invokes npx at runtime (e.g., subprocess Popen with [npx, "-y", PACKAGE]) which fetches and executes remote code provided by that package.