nano-banana-2-ecommerce-main-image

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

• Third-party content exposure detected (medium risk: 0.30). The runtime path imagegen.py skill_generate -> cmd_image -> client.generate_image sends user-supplied --prompt (free text created by the outsider) directly into the API request body as prompt without any sanitization or item selection.