omnihuman-video

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

• Third-party content exposure detected (high risk: 0.90). The skill's submit_task accepts arbitrary public image_url and audio_url (examples show https://storage.googleapis.com/... and https://example.com/...), so the agent fetches and processes untrusted third-party images and audio as part of its workflow, enabling indirect prompt injection.