firebase-ai-logic

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

• Third-party content exposure detected (high risk: 0.80). The SKILL.md explicitly mentions "Search Grounding with the built in googleSearch tool" in Core Capabilities, indicating the skill can fetch and ground on open web search results (untrusted third-party content) which the agent would read and use to influence responses and actions.