Desktop Control

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

• Third-party content exposure detected (high risk: 0.75). The skill’s runtime OCR/text ingestion path (screen.locate-text-coordinates / screen.read-all-text) reads on-screen text (via pyautogui.screenshot + easyocr.Reader.readtext), which is not authored by the operating user and can include arbitrary outsider-provided content, then returns it as readable strings in the command JSON.

MEDIUM W021: Hidden or invisible Unicode characters detected (potential obfuscation or prompt injection).

• Hidden Unicode characters detected (1 type(s) found)