nano-banana-2-seeding-image

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

• Third-party content exposure detected (medium risk: 0.30). The required runtime workflow (scripts/imagegen.py → skill_generate → cmd_image → client.generate_image) ingests the outsider-authored --prompt free text and sends it to the AI Hive generation endpoint via the prompt field in the JSON body.