comfyui-workflow-builder

Translate natural-language descriptions into executable ComfyUI workflow JSON.

86|24|Updated Feb 6, 2026
One-click install
npx skills add https://github.com/MCKRUZ/ComfyUI-Expert --skill comfyui-workflow-builder-mckruz
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: comfyui-workflow-builder
Source: https://github.com/MCKRUZ/ComfyUI-Expert/tree/main/skills/comfyui-workflow-builder
Command: npx skills add https://github.com/MCKRUZ/ComfyUI-Expert --skill comfyui-workflow-builder-mckruz

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Translates natural-language descriptions into executable ComfyUI workflow JSON.

Core Features & Use Cases

  • Generates valid ComfyUI workflow JSON from descriptive prompts, including text-to-image, image-to-video, inpainting, control nets, and LoRA stacking.
  • Validates inventory compatibility and prevents dangling connections, ensuring model/resolution/VRAM suitability.
  • Reuses references and templates to accelerate repeated workflows across projects.

Quick Start

Describe a natural-language prompt for a desired ComfyUI workflow and the skill will produce a valid, ready-to-run JSON graph.

Frequently Asked Questions about comfyui-workflow-builder

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
Does ComfyUI workflow generation validate model filenames and VRAM constraints?

Yes, you can stack LoRAs in your ComfyUI workflow by describing the desired models in your natural language prompt. The generated JSON is validated to ensure model filenames match your available inventory and VRAM constraints.

What is the best way to automate text-to-image pipeline creation in ComfyUI?

To prevent dangling inputs, the workflow generation process validates every node's class_type against your inventory. It checks for dangling connections, enforces model-resolution compatibility, and ensures VRAM suitability before outputting the JSON.

Why does my generated ComfyUI workflow fail due to missing node class_type?

Yes, you can use natural language to create inpainting and face-detail ComfyUI pipelines. The generation process translates your descriptive prompt into valid workflow JSON, applying hardware-aware validation to ensure all required nodes exist.

How do I generate ComfyUI workflow JSON from a natural language description?

Generated ComfyUI workflows fail due to missing node class_type when the validation layer detects nodes not present in your inventory. The tool enforces hardware-aware validation to ensure every class_type exists and avoids dangling inputs.