go-adk

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly shows and instructs adding the geminitool.GoogleSearch tool in SKILL.md and the quickstart snippet (and references "Google Search Grounding" in references/llms-index.md), so the agent will fetch and read public web/search results (untrusted third‑party content) that can materially influence tool routing and decisions.