concept-art

Pass

Audited by Gen Agent Trust Hub on Apr 7, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is purely instructional and provides templates for interacting with a ComfyUI instance via its API. It does not include any executable scripts, external package dependencies, or hidden commands.
  • [COMMAND_EXECUTION]: Provides a benign health check command curl ${COMFYUI_BASE_URL}/system_stats in the troubleshooting section for verifying the availability of the local ComfyUI service.
  • [DATA_EXPOSURE]: Uses an environment variable ${COMFYUI_BASE_URL} to define the target API endpoint, defaulting to a local address (http://localhost:8188). This is a standard configuration practice and does not expose sensitive credentials.
  • [INDIRECT_PROMPT_INJECTION]: While the skill describes processing 'design briefs' from external sources, it functions as a set of stylistic and workflow guidelines for the agent rather than a script that unsafely interpolates untrusted data into critical execution paths.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 7, 2026, 02:28 AM