python-sdk

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly enables ingestion of public, untrusted content—e.g., the RAG pattern using an app_tool "tavily/search-assistant@latest", internal_tools().web_search(True), and file handling that accepts remote URLs—so agents are expected to fetch and read arbitrary third‑party web pages/files.