ai-video-generation

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

• Third-party content exposure detected (high risk: 0.80). The SKILL.md shows the agent invoking inference.sh apps with arbitrary public media URLs (e.g., "image_url": "https://portrait.jpg", "video_url": "https://silent-video.mp4", arrays of https video URLs) so the system will fetch and process untrusted third‑party media as part of its workflow, which could embed instructions that influence downstream tool usage.