nano-banana-text-to-image

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

• Third-party content exposure detected (medium risk: 0.30). Outsider-authored free text is ingested directly from the user-supplied --prompt (and optional --system) into the generate/cmd_image path, which sends it to the AI Hive endpoint via chat/text/generation/image in client.generate_image(... prompt=prompt ...).