qwen-image-2
Pass
Audited by Gen Agent Trust Hub on Sep 2, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill facilitates image generation and editing by providing instructions to execute the
beltCLI tool using the Bash tool.- [EXTERNAL_DOWNLOADS]: The skill references external resources for installation and model execution, including guides on GitHub and image assets from inference.sh. These are associated with the service provider.- [INDIRECT_PROMPT_INJECTION]: The skill includes a surface for indirect prompt injection as it processes user-provided prompts and image URLs for model input. - Ingestion points: User-supplied
promptstrings andreference_imagesURIs provided in the commands. - Boundary markers: The skill uses JSON formatting for input parameters to the
beltCLI, providing structural separation between the command and the data. - Capability inventory: The skill uses the
Bashtool to run thebeltCLI utility, which performs network requests to the inference.sh API. - Sanitization: No explicit sanitization or validation logic is defined within the skill instructions for the user-provided prompt or image links.
Audit Metadata