open-interpreter

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

• Third-party content exposure detected (high risk: 0.90). This skill explicitly captures and OCRs the user's screen (oi_screenshot.py, oi_find_text.py) and runs an autonomous screenshot→LLM→action loop in OS Mode (oi_os_mode.py / SKILL.md), so arbitrary untrusted webpages or user-generated on-screen content can be read by the agent and materially influence subsequent automated actions.