scenario-gemini-image

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

• Third-party content exposure detected (low risk: 0.10). The skill runtime path described here (model_run with the required prompt plus optional useGoogleSearch grounding) relies on provider/user-selected prompt text and, if enabled, live web context; however the file does not specify a queue/feed ingestion path where an outsider can POST free text that the workflow reads without the user first selecting a specific item.