case-study-writing

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

• Third-party content exposure detected (high risk: 0.85). The required workflow invokes tavily/search-assistant, exa/search, and exa/answer with free-form queries; those tools fetch and return public web content at runtime, which the agent then includes as readable text in the LLM context, creating an indirect prompt-injection risk from outsider-authored web pages.