handdrawn-live-video-generator
Pass
Audited by Gen Agent Trust Hub on Nov 2, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides structured instructions for a creative video prompt generation task. It includes specific constraints on visual style, duration, and safety (e.g., prohibiting horror or jump scares) which are consistent with its stated creative purpose.
- [PROMPT_INJECTION]: No evidence of prompt injection or safety bypass attempts was found. The instructions use 'IMPORTANT' or '禁止' (Forbidden) only to enforce creative consistency and non-horror output, not to override system safety protocols.
- [DATA_EXFILTRATION]: No sensitive file access or network operations were detected. The skill operates entirely within the conversation context to generate text prompts.
- [REMOTE_CODE_EXECUTION]: The skill does not perform any remote code downloads or execution. It references 'MiniMax H3' as a target generation model, which is a legitimate part of the platform's video generation workflow.
- [OBFUSCATION]: No obfuscated content, encoded strings, or hidden characters were found in the instructions or metadata.
Audit Metadata