qwen-image-2

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 documentation and configuration files hosted on the official GitHub repository for inference.sh (github.com/inference-sh). These are trusted resources used for installation and guidance.
  • [COMMAND_EXECUTION]: The skill utilizes the belt CLI tool to interact with remote AI models. Execution is strictly scoped within the environment configuration to allow only the belt command, adhering to the principle of least privilege.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied text prompts and image URIs to be sent to the remote model.
  • Ingestion points: User input fields for prompt and reference_images in SKILL.md examples.
  • Boundary markers: None explicitly defined in the prompts, though the inputs are structured as JSON for the CLI tool.
  • Capability inventory: Limited to interacting with the remote image generation API via the belt tool.
  • Sanitization: Standard validation is expected to be handled by the belt CLI and the remote model provider's safety filters.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 28, 2026, 01:13 PM
Security Audit — agent-trust-hub — qwen-image-2