hive-debugger

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

• Third-party content exposure detected (high risk: 0.80). The skill inspects runtime logs, tool calls, and session memory (e.g., query_runtime_log_raw examining web_search results and get_agent_session_memory for agents like "twitter_outreach"/twitter_handles), which means it consumes and analyzes content originating from public web/social sources embedded in agent logs—untrusted third-party content that could carry indirect prompt injection.