alicloud-ai-video-retalk-test

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

• Third-party content exposure detected (high risk: 0.70). This skill's SKILL.md executable example shows it accepts arbitrary external media URLs via --video-url and --audio-url, so the agent will ingest untrusted public audio/video (user-provided web content) as part of its workflow and that media could contain instructions that materially influence processing or subsequent actions.