gemini-search

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

• Third-party content exposure detected (high risk: 1.00). This skill runs web searches via the Google Gemini CLI (SKILL.md: gemini --prompt "WebSearch: <query>") and returns the model's web-search results for the agent to read, thereby ingesting open/public third‑party content that could contain indirect prompt injection.