image-2-product-detail-page

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

• Third-party content exposure detected (medium risk: 0.30). In scripts/imagegen.py, the required runtime path generate forwards --prompt (and optional --image uploads) into client.generate_image(...prompt=prompt...) which is sent as the LLM generation input via POST /openapi/v1/generation/image; this is outsider-authored free text provided at runtime without any prior item selection.