video-analyzer
Warn
Audited by Snyk on Feb 24, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (high risk: 0.90). SKILL.md and README explicitly instruct using yt-dlp to download videos from public sites (B站/YouTube) and then upload those untrusted, user-generated videos to the Responses API for the agent to analyze, so third-party video content can directly influence the agent's outputs and actions.
Audit Metadata