The Agent Skills Directory

[SAFE]: The skill provides legitimate documentation and usage examples for interacting with AI video generation models via the belt command-line interface.
[EXTERNAL_DOWNLOADS]: The skill references external resources including documentation and installation scripts located at github.com/inference-sh and inference.sh. These resources originate from the skill's official vendor and are used for setup and instructional purposes.
[COMMAND_EXECUTION]: Instructions involve using the belt CLI tool for media generation. The tool usage is constrained to the vendor's application ecosystem and matches the stated purpose of the skill.
[PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection as it processes external data such as prompts and media URLs through the belt CLI tool in SKILL.md. No boundary markers or specific sanitization routines are defined in the instructions to delimit untrusted input. However, this capability is core to the primary purpose of AI video generation and does not represent a malicious intent.

ai-video-generation