art

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly enables real-time web search grounding (see nano-banana-guide.md "Web Search Grounding" and the CLI example using --grounded / API tools: [{ googleSearch: {} }]), meaning the agent fetches and uses public web pages/URLs to ground image generation — exposing it to untrusted third-party content that can materially influence prompts and generation behavior.