seedance-v2

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly accepts arbitrary image_url/video_url/audio_url references that the RunComfy model server fetches and incorporates into generation (see "How it works" and "Third-party content" in SKILL.md), so untrusted user-provided web media can influence model behavior and enable indirect prompt injection.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.70). The skill calls the RunComfy Model API at runtime (POST to https://model-api.runcomfy.net/v1/models/bytedance/seedance-v2/pro), which executes remote model code to produce outputs and thus is a required external runtime dependency that controls the agent's behavior.