media-generation

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly accepts arbitrary external image URLs (e.g., the Image prompts in skincare-promo.tsx and the reference-based example using "https://reference-photo.jpg"/"https://reference-photo-url.jpg"), which the runtime fetches and uses as reference input for generation, so untrusted third-party content would be ingested and interpreted during rendering.