visual-design-system-extractor

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

• Third-party content exposure detected (low risk: 0.10). The only outsider-authored free text the runtime clearly ingests is the user’s prompt/command input (e.g., extract <refs> and fonts <brief>), which can steer what reference images/URLs to use and thus indirectly affect what the model “reads” (via user-supplied references), without any documented ingestion of watched queue/chat/ticket/email body text before selecting a specific item.