gpt-image

Pass

Audited by Gen Agent Trust Hub on Sep 2, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references installation instructions and resources from the official inference-sh GitHub repository and the inference.sh domain.
  • [COMMAND_EXECUTION]: The skill is configured to use the belt CLI tool specifically for running image generation models.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data through text prompts and image URLs. Ingestion points: 'prompt', 'images', and 'mask' parameters in SKILL.md. Boundary markers: None. Capability inventory: Execution of 'belt' CLI commands as defined in allowed-tools. Sanitization: None present. The risk is considered standard for the skill's primary function of image generation.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 2, 2026, 02:20 PM
Security Audit — agent-trust-hub — gpt-image