ai-image-generation
Pass
Audited by Gen Agent Trust Hub on Nov 18, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill references installation instructions hosted on the inference-sh GitHub repository (
raw.githubusercontent.com/inference-sh/skills/...). This is a legitimate resource related to the skill's primary vendor. - [COMMAND_EXECUTION]: The skill utilizes the
beltCLI tool via the allowedBash(belt *)tool. This is the intended mechanism for interacting with the image generation service. - [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied text prompts for image generation, creating a surface for potential injection.
- Ingestion points: User prompts passed to the
--inputargument of thebelt app runcommand. - Boundary markers: Examples use JSON formatting (e.g.,
'{"prompt": "..."}') to provide structural separation for user input. - Capability inventory: The skill executes commands via the
beltCLI tool. - Sanitization: No explicit sanitization or escaping logic is described in the markdown instructions; the agent relies on the CLI's handling of the JSON payload.
Audit Metadata