SlavaSextonSlavaSextonCommunityยท1 Agent Skills Included

ComfyUI-Agent-Kit

Local ComfyUI image, video, and audio generation control

Drives a local ComfyUI install to generate images, video, and audio, build and validate workflows, and manage models on your own GPU. Eliminates manual graph wiring, wrong model picks for your VRAM, and per-generation cloud fees. Ships per-model prompt recipes, a 581-template library, and hardware-aware model selection so every job runs right the first time.
npx skills add SlavaSexton/ComfyUI-Agent-Kit --all -g -y
Available:

Explains how each supported coding agent (Claude Code, Codex, Gemini CLI, Qwen Code) connects to the shared ComfyUI MCP server and where its knowledge file is installed.

All Skills in This Repository (1)

Pure Emerald Level Indicators

Frequently Asked Questions

FAQPage Schema
How to install ComfyUI-Agent-Kit?โ–ผ

Run `npx skills add SlavaSexton/ComfyUI-Agent-Kit --all -g -y` in your terminal to install all skills in this kit globally.

What does ComfyUI-Agent-Kit do?โ–ผ

It lets your coding agent drive a local ComfyUI install to generate images, video, and audio, build and validate workflows, and download models, all on your own hardware with no cloud account.

Which agents work with ComfyUI-Agent-Kit?โ–ผ

It supports Claude Code, Codex, Gemini CLI, and Qwen Code through one shared core plus a thin adapter per agent, and GLM via Claude Code is covered too.

Does it pick the right model for my GPU?โ–ผ

Yes. It detects your VRAM, RAM, and free disk, then recommends the model variant that fits and refuses downloads that will not fit before wasting bandwidth.

Do I need ComfyUI experience to use it?โ–ผ

No. The agent uses a 581-template library and per-model prompt recipes to build and run workflows from your plain-English request, and can even start ComfyUI for you.

Related Repositories in Design & Creative

View All in Design & Creativeโ†’