nano-banana-pro

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

• Third-party content exposure detected (high risk: 0.70). The skill runs scripts/generate_image.py, which sends the user-provided --prompt to the Gemini API and then prints any part.text from the API response into the agent/LLM context; this is outsider-authored free text coming from a third-party model response at runtime.