bytedance-seedance-2-0

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

• Third-party content exposure detected (high risk: 0.80). The skill's image-to-video workflow explicitly requires publicly reachable HTTPS image_url/end_image_url inputs (see SKILL.md and references/MODELS.md), so the model ingests third-party user-provided web content that can materially influence generation and could carry embedded instructions for indirect prompt injection.