qwen-image-2

Pass

Audited by Gen Agent Trust Hub on Aug 7, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references official documentation and installation scripts hosted on the 'inference-sh' GitHub repository to facilitate setup.- [COMMAND_EXECUTION]: The skill operates by invoking the 'belt' CLI tool. The execution environment is protected by an 'allowed-tools' configuration that limits the Bash tool to commands starting with 'belt', preventing arbitrary code execution.- [PROMPT_INJECTION]: The skill processes user-defined inputs for image generation, creating a potential surface for indirect prompt injection.
  • Ingestion points: User-provided 'prompt' and 'reference_images' within the belt CLI command.
  • Boundary markers: None identified in the skill instructions to separate instructions from user data.
  • Capability inventory: Access is restricted to the 'belt' CLI via the Bash tool.
  • Sanitization: No input sanitization or validation logic is specified within the skill body.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 7, 2026, 06:33 PM
Security Audit — agent-trust-hub — qwen-image-2