seo-content-brief

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

• Third-party content exposure detected (low risk: 0.10). The required workflow runs web searches (tavily/search-assistant, exa/search) and extracts full page text from top-ranking URLs (tavily/extract), so outsider-authored web content enters the LLM context, but only after the agent issues a search or selects specific URLs, so an attacker can publish poison but cannot force delivery.