image-upscaling

Pass

Audited by Gen Agent Trust Hub on Aug 28, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references installation documentation and assets hosted on the inference-sh GitHub organization and the official inference.sh domain. These references are used to set up the necessary command-line environment.
  • [COMMAND_EXECUTION]: The skill is configured to use the Bash tool with the belt command to perform image upscaling operations. It includes examples of logging into the service and running specific AI models with user-provided parameters.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests external image URLs provided by the user. While this presents an injection surface typical for media processing tasks, the risk is mitigated by passing inputs as structured JSON arguments to the belt CLI.
  • [CREDENTIALS_SAFE]: The documentation includes instructions for managing service authentication using the belt login command, which is a standard procedure for CLI-based service interaction.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 28, 2026, 01:13 PM
Security Audit — agent-trust-hub — image-upscaling