ai-ecommerce-expert-brand-content-consistency

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

• Third-party content exposure detected (medium risk: 0.30). Required runtime workflow routes user-authored parameters into the IMIVA MCP call (e.g., via create_visual_migration_task arguments like subjectDescriptions and referenceImage/productImages), so outsider free text can be ingested through the tool input without any forced selection step beyond crafting the request.

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

• Potentially malicious external URL detected (high risk: 1.00). The skill invokes npx to run the remote npm package @infimind/ecom-content-cli@latest at runtime (see subprocess call), which will fetch and execute remote code from that package.