qwencloud-video-generation

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly accepts and passes arbitrary external HTTP(S) URLs (e.g., img_url, first_frame_url, last_frame_url, reference_urls, audio_url, video_url) from users to the video-generation API — see SKILL.md and references/request-fields.md and the examples — meaning untrusted third-party media/URLs are ingested as part of the workflow and can materially influence mode selection, payloads, and generated outputs (i.e., enabling indirect prompt injection).