tavily-tools

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

• Third-party content exposure detected (high risk: 1.00). The skill explicitly fetches and ingests arbitrary public web content via commands like extract --urls, crawl --url, map --url, and search (which return results from the open web), so the agent will read and interpret untrusted, user-generated third-party pages and search results, enabling indirect prompt injection.