comfy-imagegen
Installation
SKILL.md
comfy-imagegen
Use this skill for image generation, image editing, and image upscaling through
the comfy-imagegen CLI. Local modes use models under /mnt/models/comfyui.
If a supported built-in model is missing, use comfy-model-downloader to fetch
only the requested capability before running inference.
The CLI is quiet by default and prints only final JSON. Use --verbose only when
debugging ComfyUI runtime output, warnings, or progress bars.
If comfy-imagegen or comfy-models is not available, use
comfy-tools-setup first. In this repository, prefer uv run comfy-imagegen;
outside the repo, let comfy-tools-setup install the CLIs with uv tool.
At the start of every image workflow, start or reuse the local Comfy Media gallery for the active output directory: