nano-banana-build

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly instructs the agent to fetch and inspect public web content (e.g., raw.githubusercontent.com URLs in references/source_code.md) and to perform Google Search grounding (references/recipes.md), which the agent is expected to read and use in generation, exposing it to untrusted third-party content.