artokunCommunity·42 Agent Skills Included

comfyui-mcp

Control ComfyUI image and video generation with any AI model

Generates images, video, and audio in ComfyUI through natural language commands from any LLM. Builds and edits live workflow graphs, downloads models, manages custom nodes, and fixes VRAM or sampler errors automatically. Ships 42 ready skills for Flux, WAN, LTX, Qwen, Z-Image, and more, eliminating manual node wiring and parameter guesswork.

715|111

↗

npx skills add artokun/comfyui-mcp --all -g -y

All Skills in This Repository (42)

Pure Emerald Level Indicators

📦 In Repo

artokun

director

Orchestrate a full short film production pipeline from a text story into a final video.

wan-flf-video

Generate First-Last-Frame videos with WAN 2.2 I2V models.

ltxv2-video

Generate videos from text or images using the LTX-V2 19B model.

flux-txt2img

Generate images with Flux diffusion models and configure KSampler settings.

wan-t2v-video

Generate videos from text prompts using dual-model UNETs and CLIP encoders.

prompt-engineering

Explain ComfyUI prompt syntax, weight modifiers, and model-specific strategies.

z-image-txt2img

Generate images with Z-Image models in ComfyUI using LoRAs and ControlNet.

comfyui-core

Explain ComfyUI workflow JSON formats, node schemas, and MCP tool usage.

troubleshooting

Diagnose ComfyUI errors with debugging commands and targeted fixes.

model-compatibility

Detail ComfyUI model compatibility requirements for SD 1.5, SDXL, Flux, SD3, and LTXV.

qwen-txt2img

Generate images with the Qwen Image 2512 model using integrated or separate workflows.

qwen-image-edit

Edit images with Qwen vision-language models via ComfyUI workflows.

Community

Advanced

Frequently Asked Questions

FAQPage Schema

How to install comfyui-mcp?▼

Run `npx skills add artokun/comfyui-mcp --all -g -y` in your terminal to install the full skill suite globally.

What does comfyui-mcp do?▼

It lets an AI agent operate ComfyUI for you: generating images and video, building and editing workflow graphs, downloading models, and managing custom nodes from plain-English requests.

Which AI models work with comfyui-mcp?▼

Claude, ChatGPT, Gemini, and any OpenAI-compatible or local model via Ollama all work, including fully offline free local models.

Can comfyui-mcp generate video, not just images?▼

Yes. It includes skills for WAN 2.2, LTX-2.3, MiniMax H3, video extension, upscaling, and talking-avatar lip-sync pipelines.

Do I need ComfyUI experience to use it?▼

No. The agent picks the right models, samplers, and settings for each task, so you describe what you want and it handles the technical workflow.