The Agent Skills Directory

[SAFE]: The skill provides legitimate instructions for video generation and does not contain any malicious patterns, obfuscation, or unauthorized data access attempts. All external resources belong to the verified vendor infrastructure.
[COMMAND_EXECUTION]: The skill utilizes the belt CLI tool for interacting with video models. The execution environment is restricted to this specific tool via the allowed-tools frontmatter configuration, which is a security best practice to limit the agent's attack surface.
[EXTERNAL_DOWNLOADS]: The skill references documentation, images, and installation scripts hosted on official vendor domains (inference.sh) and the official inference-sh GitHub repository. These are documented neutrally as functional requirements.
[REMOTE_CODE_EXECUTION]: The skill suggests using npx to add related skills from the vendor's repository. This represents a standard workflow for extending agent capabilities with verified packages from the vendor's own collection.

google-veo