happyhorse
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill references external installation instructions and documentation hosted on the official GitHub repository for inference-sh. These references are used to guide the user through the setup process for the required CLI tool.
- [INDIRECT_PROMPT_INJECTION]: The skill involves processing user-provided text prompts and URLs for images and videos to generate media content. While this creates a surface for indirect prompt injection, the risk is mitigated by the execution environment which is strictly scoped to the 'belt' CLI tool.
- Ingestion points: User-provided values for 'prompt', 'first_frame', 'video', and 'reference_images' in the model execution commands.
- Boundary markers: None explicitly defined in the skill body.
- Capability inventory: The skill is restricted to using the Bash tool for the 'belt' CLI exclusively.
- Sanitization: No explicit sanitization or filtering is described in the skill documentation.
Audit Metadata