creating-an-agent

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly includes WebSearch/WebFetch in the Tool Selection and the "Research Agent" example which instructs the agent to WebSearch, WebFetch, extract and synthesize information from web sources, so it will read and interpret untrusted public web content (third-party sites/results).