qwen-image-2-1-creator
Pass
Audited by Gen Agent Trust Hub on Nov 10, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill provides local Python scripts (
audit_comfyui.py,inject_comfyui.py,validate_task.py,video_handoff.py) that use only the Python standard library for JSON manipulation and validation. Besides basic file I/O for reading/writing the JSON manifests and workflows specified by the user, no dangerous system operations were found.- [SAFE]: No unauthorized network calls or remote data fetching mechanisms were identified in the script logic. External model references (such as Wan or MiniMax) are handled purely as metadata for handoff files rather than active execution components.- [SAFE]: The skill includes a dedicated auditing script (audit_comfyui.py) designed to prevent unauthorized modifications to ComfyUI workflow files, allowing only specific prompt nodes to be updated while verifying that the rest of the workflow structure (links, nodes, metadata) remains unchanged.- [SAFE]: Documentation and instructions emphasize transparency, requiring human-in-the-loop review and explicitly stating that external video models and Prompt Enhancer (PE) weights are not provided by the skill and must be provisioned separately by the user.
Audit Metadata