How do I install the comfyui mcp server?

Use a recorded configuration for your client, replace example credential values and paths, and verify the documented runtime. Restart or reconnect the client, then inspect the available tools.

What credentials does it need?

It reads `CIVITAI_API_TOKEN`, `COMFYUI_API_KEY`, `COMFYUI_MCP_HTTP_TOKEN`, `COMFYUI_URL` from the environment. Store these in your client's env block or a secrets manager — never commit them to a repository — and grant the narrowest scope that still covers your use case.

SSyncDev

Start a project

Comfyui MCP Server

Local-first, agent-native control plane for ComfyUI — MCP server + autonomous sidebar agent that drives your live graph in natural language on ANY

Remote serverstreamable-httpPython

What is the Comfyui MCP Server?

Local-first, agent-native control plane for ComfyUI — MCP server + autonomous sidebar agent that drives your live graph in natural language on ANY LLM

What the server does

One-click GPU pod — a ready-to-run ComfyUI with this project + Agent Panel + ComfyUI-Manager v2 preinstalled, on your own GPU. No setup.

More MCP options

Compare MCP servers by category, or browse hosted MCP servers if you prefer an endpoint over a local process.

Recorded tools

Check the linked publisher documentation and your connected client's tool list. Names and availability can vary with the server version and account permissions.

Tool	What it does
queue	One action-parameterized tool for the execution queue: list (running + pending), status (one job by prompt ID), get_workflow (a pending job's full payload), mov
get_system_stats	Get system info — GPU, VRAM, Python version, OS
visualize_workflow	One action-parameterized tool for rendering and converting a workflow you pass in: render (Mermaid flowchart, nodes grouped by category), render_hierarchical (t
create_workflow	One action-parameterized tool for authoring: create (from templates: txt2img, img2img, upscale, inpaint, controlnet, ip_adapter, ace_step_15, stable_audio_3), m
get_workflow	One action-parameterized tool for READING a saved workflow file: list (the user library, subfolders included), get (one workflow's JSON by filename), analyze (a
save_workflow	One action-parameterized tool for WRITING to the library: save (store a workflow — overwrites a same-filename file), lock (record a provenance lock: SHA-256 per
upload_image	Copy a local image into ComfyUI's input/ directory for img2img, inpaint, or ControlNet
download_model	Find and fetch model weights, and track the transfers. action: download (from a URL, into the correct ComfyUI subdirectory), status, cancel, search (HuggingFace
list_local_models	What is installed, and where ComfyUI looks. action: list (installed models by type: checkpoints, loras, vae, upscale_models, controlnet, embeddings, clip, unet,
clear_vram	Free GPU VRAM by unloading cached models — calls ComfyUI's /free endpoint, reports before/after stats
search_custom_nodes	Search the ComfyUI Registry for node packs by keyword (action: "search"), or get one pack's full details (action: "details")
comfy_cli	One action-parameterized tool for the whole official CLI: status, server_start/server_stop/server_restart, jobs_list/jobs_status/jobs_wait/jobs_watch/jobs_cance
kitchen	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 (apply) — a failed probe is un

How to install the Comfyui MCP Server

{
  "mcpServers": {
    "comfyui": {
      "command": "npx",
      "args": [
        "-y",
        "comfyui-mcp"
      ],
      "env": {
        "CIVITAI_API_TOKEN": "YOUR_CIVITAI_API_TOKEN"
      }
    }
  }
}

From the project documentation. Replace placeholders and example paths before connecting.

Configuration

Variable	Description	Required
CIVITAI_API_TOKEN	Declared in the maintainer’s client configuration.	Check documentation

Example prompts to try

“Use Comfyui to Agents.”
“Use Comfyui to Hooks.”
“Use Comfyui to Defaults.”

Frequently asked questions

Local-first, agent-native control plane for ComfyUI — MCP server + autonomous sidebar agent that drives your live graph in natural language on ANY LLM Confirm the tools offered by your installed version in the connected client. Tool availability and permissions depend on the server and account configuration.