hive-concepts

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

• Third-party content exposure detected (high risk: 1.00). This skill explicitly instructs event_loop nodes (e.g., the "search-web" NodeSpec and the "research_node" examples) to use tools like "web_search" and "web_scrape" to fetch and extract results from the open web, meaning the agent will read and interpret untrusted public third-party content.