image-upscaling

Pass

Audited by Gen Agent Trust Hub on Sep 2, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill fetches installation scripts and documentation from the official inference-sh GitHub repository, which is necessary for the skill's primary function of interfacing with the upscaling service.
  • [COMMAND_EXECUTION]: The skill employs the belt CLI to execute image processing tasks on remote infrastructure. This is the intended behavior for an AI agent skill that wraps a command-line interface.
  • [INDIRECT_PROMPT_INJECTION]: The skill accepts external image URLs as input for upscaling. This represents a potential surface for indirect prompt injection if the agent processes malicious instructions embedded within the image data or its associated metadata.
  • Ingestion points: Ingests image URLs via the --input flag in shell commands (e.g., SKILL.md).
  • Boundary markers: None present in example commands.
  • Capability inventory: Executes shell commands via the belt CLI.
  • Sanitization: No explicit sanitization of input URLs is performed within the instruction set.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 2, 2026, 02:23 PM
Security Audit — agent-trust-hub — image-upscaling