mv-storyboard-director
Pass
Audited by Gen Agent Trust Hub on Aug 12, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill consists of markdown-based instructions for creative storyboard design. It does not contain any scripts, binaries, or commands that could be executed in a shell or runtime environment.
- [DATA_EXPOSURE]: There are no hardcoded credentials, sensitive file paths, or network exfiltration patterns. The skill explicitly restricts the agent from calling APIs or interacting with external video generation platforms.
- [REMOTE_CODE_EXECUTION]: No remote script downloads or package installations are performed. The skill operates purely within the text-based prompt-response cycle.
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied song lyrics and audio references. While it does not implement specific sanitization or delimiters for this untrusted data (Ingestion points: SKILL.md), it possesses no dangerous capabilities (Subprocess, Network, File-write) that could be abused if an injection occurred. The attack surface is minimal.
Audit Metadata