ai-image-generation
Pass
Audited by Gen Agent Trust Hub on May 20, 2026
Risk Level: SAFE
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill references installation documentation hosted on a public GitHub repository (raw.githubusercontent.com) associated with the inference.sh service.
- [COMMAND_EXECUTION]: Provides example commands for the 'belt' CLI tool, such as 'belt login' and 'belt app run', which are the primary intended methods for interacting with the image generation models.
- [INDIRECT_PROMPT_INJECTION]: The skill accepts user-supplied text prompts to generate images. This is a standard functional surface where untrusted data (the prompt) is processed by an external model. The risk is mitigated by the structured JSON input format and the specific scope of the 'belt' tool.
Audit Metadata