understand-image-video

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

• Third-party content exposure detected (medium risk: 0.30). The runtime ingests outsider-authored free text when a user provides an arbitrary --prompt (or stdin) that is sent as user_text to https://apihub.agnes-ai.com/v1/chat/completions in scripts/agnes-understand.py, so the agent reads attacker-controlled free text without needing a prior specific item selection.