ai-image-generation

Pass

Audited by Gen Agent Trust Hub on Sep 2, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references external installation instructions at https://raw.githubusercontent.com/inference-sh/skills/refs/heads/main/cli-install.md and provides links to the inference.sh documentation. These resources are hosted on infrastructure associated with the skill vendor.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes user-provided strings (prompts) and external image URLs to perform image generation and editing tasks. This represents a potential attack surface where adversarial input could influence the behavior of the underlying AI models or the CLI tool.
  • Ingestion points: User-supplied prompts and image URLs included in the JSON input for belt app run commands (e.g., in SKILL.md).
  • Boundary markers: The skill uses JSON formatting to structure the input arguments for the CLI, providing a basic level of data separation.
  • Capability inventory: The skill utilizes the Bash tool to execute belt CLI commands.
  • Sanitization: No explicit validation or escaping logic for user prompts or image metadata is described in the skill instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 2, 2026, 02:18 PM
Security Audit — agent-trust-hub — ai-image-generation