> ## Documentation Index
> Fetch the complete documentation index at: https://comfyui-mcp.artokun.io/docs/llms.txt
> Use this file to discover all available pages before exploring further.

# ComfyUI MCP

> The local-first, agent-native control plane for ComfyUI — an MCP server + Claude Code plugin that generates images, video and audio, authors and runs workflows, and manages models and custom nodes in natural language. Local, remote, or Comfy Cloud.

**ComfyUI MCP** is the **local-first, agent-native control plane** for
[ComfyUI](https://github.com/comfyanonymous/ComfyUI) — an [MCP](https://modelcontextprotocol.io)
server and Claude Code plugin. It connects your AI assistant to a running ComfyUI instance —
**local**, **remote**, or **[Comfy Cloud](https://cloud.comfy.org)** — so you can generate images,
video and audio, execute and author workflows, manage models and custom nodes, and control the
server, all in natural language. Not a thin prompt-to-image bridge — a full agent that operates
ComfyUI for you.

<Frame caption="The Agent panel driving ComfyUI end to end — reading what's installed locally, wiring the graph, freeing VRAM, and running the render">
  <video autoPlay muted loop playsInline controls poster="./images/demo-poster.jpg" className="w-full rounded-xl" src="https://mintcdn.com/artokun/Me1yID9TLGaEqCsU/videos/comfyui-mcp-demo.mp4?fit=max&auto=format&n=Me1yID9TLGaEqCsU&q=85&s=112cce3981a4847ec9cf93cd89355f58" data-path="videos/comfyui-mcp-demo.mp4" />
</Frame>

<CardGroup cols={2}>
  <Card title="Quickstart" icon="rocket" href="./quickstart">
    Add the server to Claude Code and generate your first image in \~2 minutes.
  </Card>

  <Card title="Tool Reference" icon="wrench" href="./tools/image-generation">
    38 tools, auto-documented from their live schemas.
  </Card>

  <Card title="Installation" icon="download" href="./installation">
    Install ComfyUI and wire up the MCP server (local or remote).
  </Card>

  <Card title="Configuration" icon="gear" href="./configuration">
    Environment variables, transports, tokens, and remote targeting.
  </Card>
</CardGroup>

## What you can do

<CardGroup cols={3}>
  <Card title="Generate" icon="image">
    Text-to-image with `generate_image`, plus ControlNet and IP-Adapter conditioning.
  </Card>

  <Card title="Author workflows" icon="pen-ruler">
    Create from templates, modify, validate, visualize, and round-trip a compact DSL.
  </Card>

  <Card title="Manage models" icon="box">
    Search HuggingFace/CivitAI, download, list, and remove models.
  </Card>

  <Card title="Custom nodes" icon="puzzle">
    Install, update, snapshot, and bisect node packs via ComfyUI-Manager.
  </Card>

  <Card title="Install & maintain" icon="terminal">
    Install/update ComfyUI, inspect the environment, start/stop the server.
  </Card>

  <Card title="Local, remote, or cloud" icon="globe">
    Auto-detects a local install, points at a remote with `--comfyui-url`, or talks to
    [Comfy Cloud](https://cloud.comfy.org) via `COMFYUI_API_KEY` — no local ComfyUI needed.
  </Card>
</CardGroup>

<Note>
  Works on **macOS**, **Linux**, and **Windows**, and auto-detects a local ComfyUI installation
  and port. Source and issues live on [GitHub](https://github.com/artokun/comfyui-mcp).
</Note>

## Scope: local, remote, or Comfy Cloud

`comfyui-mcp` is the community MCP for **local** and **remote** ComfyUI — that's the primary
target (Mac/Linux/Windows installs, RunPod, VPS, LAN, etc.).

For **Comfy Cloud** users, Comfy-Org ships [official agent tooling](https://docs.comfy.org/agent-tools) — Comfy Cloud MCP (public beta) and the Comfy In-App Agent (private alpha) — maintained by the Comfy team and running on Comfy Cloud. If you have no GPU or want zero setup, use that ([side-by-side](./local-vs-comfy-cloud)). `comfyui-mcp` also includes a community cloud-mode (`COMFYUI_API_KEY`), so a single MCP can target all three deployment shapes from one config — pick whichever fits your workflow.

***

## Maintainer

Built and maintained by [**@artokun**](https://github.com/artokun) — regular contributor to
[Comfy-Org/ComfyUI\_frontend](https://github.com/Comfy-Org/ComfyUI_frontend/pulls?q=is%3Apr+author%3Aartokun)
(v2 graph renderer / subgraph rendering / promoted widgets) and the
[ComfyUI Python core](https://github.com/Comfy-Org/ComfyUI/pulls?q=is%3Apr+author%3Aartokun).

**Comfy-Org folks** (or anyone hiring around the ComfyUI ecosystem): [art.longbottom.jr@gmail.com](mailto:art.longbottom.jr@gmail.com).