nano-banana-pro-ad-image

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

• Third-party content exposure detected (medium risk: 0.30). The runtime path scripts/imagegen.py ingests outsider-provided free text directly as --prompt (and optional --system) and forwards it to the AI Hive endpoint via AiHiveClient.generate_image(..., prompt=...) / chat_text(..., messages=...) without any selection step.