unbox-ai

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

• Third-party content exposure detected (low risk: 0.10). The skill analyzes local AI agent trace files and devtools databases, which contain message histories and tool outputs from prior agent runs; these files may include untrusted external data (such as user inputs or web content ingested during those runs), but the agent must explicitly choose which trace or file to inspect rather than monitoring an unselected external feed.