research

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

• Third-party content exposure detected (high risk: 0.90). The skill sends arbitrary research queries to Tavily's MCP endpoint (https://mcp.tavily.com/mcp) which performs "automatic source gathering" from web/public sources and returns third‑party results that the agent reads and synthesizes, exposing it to untrusted open-web/user-generated content.