nano-banana

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

• Third-party content exposure detected (high risk: 0.90). The skill includes explicit Google Search grounding (e.g., the "Advanced Features" example with tools=[{"google_search": {}}] and the google_search flag in gemini_images._build_config), which lets the agent fetch and consume real-time public web content for generation—exposing it to untrusted third‑party data that could carry indirect prompt injections.