gemini-cli

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

• Third-party content exposure detected (high risk: 1.00). The skill explicitly uses real-time web search and URL fetching (e.g., tools.md's "google_web_search" and "web_fetch" and templates.md examples like "Fetch and summarize https://example.com/docs"), so the agent will fetch and read untrusted public web content.