image-upscaling
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to use the
beltCLI tool for image processing tasks. Command execution is limited to thebeltutility as specified in the platform's tool restrictions. - [EXTERNAL_DOWNLOADS]: The skill references external resources for installation and platform integration. It points to configuration guidelines hosted on GitHub (
raw.githubusercontent.com/inference-sh/skills/) and suggests adding supplemental skills from the inference-sh ecosystem. - [INDIRECT_PROMPT_INJECTION]: The skill handles external inputs such as image URLs and processes structured data from CLI tool outputs.
- Ingestion points: Image URLs provided in the command arguments.
- Boundary markers: No explicit delimiters or instructions are used to isolate untrusted input within the command structure.
- Capability inventory: The skill is scoped to image processing via the
beltCLI. - Sanitization: No specific input validation or escaping mechanisms are implemented in the provided examples.
Audit Metadata