vmos-edge-control-api

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly instructs the agent to fetch and interpret device UI content (e.g., GET /screenshot/format, /screenshot/raw, /screenshot/data_url and GET /accessibility/dump_compact from the device base URL http://{host_ip}:18182/android_api/v2/{db_id}) and to plan/execute clicks and inputs based on that content, so untrusted app/webpage/user-generated content shown on the device can materially influence actions.