comfyui-mcp MCP server
Local-first, agent-native control plane for ComfyUI — MCP server + sidebar agent that generates images, video & audio, authors and runs workflows, and edits your live graph in natural language on ANY LLM (Claude, ChatGPT, Gemini, offline Ollama, or any hosted model). 178 tools, 36 AI skills, 55 installer packs. Local, LAN, VPS, or Comfy Cloud.
What comfyui-mcp does
Summarised from the project README; check the repository for the latest details.
Tools exposed (15)
Defaultsqueue– One action-parameterized tool for the execution queue: `list` (running + pending), `status` (one job by prompt ID), `getworkflow` (a pendingget_system_stats– Get system info — GPU, VRAM, Python version, OSvisualize_workflow– One action-parameterized tool for rendering and converting a workflow you pass in: `render` (Mermaid flowchart, nodes grouped by category),create_workflow– One action-parameterized tool for authoring: `create` (from templates: `txt2img`, `img2img`, `upscale`, `inpaint`, `controlnet`, `ipadapter`get_workflow– One action-parameterized tool for READING a saved workflow file: `list` (the user library, subfolders included), `get` (one workflow's JSONsave_workflow– One action-parameterized tool for WRITING to the library: `save` (store a workflow — overwrites a same-filename file), `lock` (record a provupload_image– Copy a local image into ComfyUI's `input/` directory for img2img, inpaint, or ControlNetdownload_model– Find and fetch model weights, and track the transfers. `action`: `download` (from a URL, into the correct ComfyUI subdirectory), `status`, `list_local_models– What is installed, and where ComfyUI looks. `action`: `list` (installed models by type: checkpoints, loras, vae, upscalemodels, controlnet,clear_vram– Free GPU VRAM by unloading cached models — calls ComfyUI's `/free` endpoint, reports before/after statssearch_custom_nodes– Search the ComfyUI Registry for node packs by keyword (`action: "search"`), or get one pack's full details (`action: "details"`)comfy_cli– Search actual loaded node classes through official `comfy nodes search` (action:"searchnodes")Diagnosticskitchen– See what comfy-kitchen can do on this GPU (`status`), find where a graph leaves it on the table (`assess`), and apply the faster path (`appl
Requirements
- Node.js >= 22.0.0
- ComfyUI running locally
- Node.js (runs with npx)
- Environment variables: CIVITAI_API_TOKEN
Repository: artokun/comfyui-mcp
Website: https://comfyui-mcp.artokun.io/docs
Stars: ★824
Language: TypeScript
Last updated: 2026-10-14
agent-skillsai-agentclaude-codeclaude-plugincomfyuicomfyui-extensioncomfyui-mcpcomfyui-mcp-serverfluximage-generationlocal-firstmcpmcp-servermodel-context-protocolofflineollamaself-hostedstable-diffusionvideo-generationwan
Website: https://comfyui-mcp.artokun.io/docs
Stars: ★824
Language: TypeScript
Last updated: 2026-10-14
agent-skillsai-agentclaude-codeclaude-plugincomfyuicomfyui-extensioncomfyui-mcpcomfyui-mcp-serverfluximage-generationlocal-firstmcpmcp-servermodel-context-protocolofflineollamaself-hostedstable-diffusionvideo-generationwan
How to install comfyui-mcp MCP in Claude Desktop, Cursor and VS Code
Install
Taken from the project README (README mcpServers JSON). Check the README for required API keys before use.
Claude Desktop
claude_desktop_config.json
{
"mcpServers": {
"comfyui": {
"command": "npx",
"args": [
"-y",
"comfyui-mcp"
],
"env": {
"CIVITAI_API_TOKEN": ""
}
}
}
}Cursor
~/.cursor/mcp.json
{
"mcpServers": {
"comfyui": {
"command": "npx",
"args": [
"-y",
"comfyui-mcp"
],
"env": {
"CIVITAI_API_TOKEN": ""
}
}
}
}VS Code
.vscode/mcp.json
{
"servers": {
"comfyui": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"comfyui-mcp"
],
"env": {
"CIVITAI_API_TOKEN": ""
}
}
}
}Add to config generator – combine it with other servers, or get the Windsurf format.
Maintain this project? Get it featured for free during launch →
Related Memory & Knowledge MCP servers
ruflo by ruvnet · ★74702
🌊 The original agent harness. Deploy intelligent multi-player swarms, coordinate autonomous workflows, and build conversational AI systems.
Memory & Knowledge
🌊 The original agent harness. Deploy intelligent multi-player swarms, coordinate autonomous workflows, and build conversational AI systems.
Memory & Knowledge
context-mode by mksglu · ★29252
Context window optimization for AI coding agents. Sandboxes tool output (98% reduction), persists session memory, and enforces routing acr
Memory & Knowledge
Context window optimization for AI coding agents. Sandboxes tool output (98% reduction), persists session memory, and enforces routing acr
Memory & Knowledge
Trail-Code by trailhq · ★10034
Turbocharge Claude Code, Cursor, Codex, Gemini & every coding agent: faster, cheaper, with contextual understanding specific to your codebas
Memory & Knowledge
Turbocharge Claude Code, Cursor, Codex, Gemini & every coding agent: faster, cheaper, with contextual understanding specific to your codebas
Memory & Knowledge
osaurus by osaurus-ai · ★8048
Own your AI. The native macOS harness for AI agents -- any model, persistent memory, autonomous execution, cryptographic identity. Built in
Memory & Knowledge
Own your AI. The native macOS harness for AI agents -- any model, persistent memory, autonomous execution, cryptographic identity. Built in
Memory & Knowledge
semble by MinishLab · ★6209
Fast and Accurate Code Search for Agents. Uses 99% fewer tokens than grep+read
Memory & Knowledge
Fast and Accurate Code Search for Agents. Uses 99% fewer tokens than grep+read
Memory & Knowledge
cortex by cortex-docs · ★3248
Cortex - Generates interactive API documentation, typed SDKs, and MCP servers from OpenAPI, AsyncAPI, GraphQL, gRPC, OpenRPC, and Markdown.
Memory & Knowledge
Cortex - Generates interactive API documentation, typed SDKs, and MCP servers from OpenAPI, AsyncAPI, GraphQL, gRPC, OpenRPC, and Markdown.
Memory & Knowledge
Weekly MCP digest – new and trending MCP servers in your inbox.