Comfy

Provides image generation by sending prompts to a remote Comfy server through the FastMCP framework.

python

GitHub Stars

python

Language

6 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "mcp-mirror-lalanikarim_comfy-mcp-server": {
      "command": "python",
      "args": [
        "comfy-mcp-server.py"
      ],
      "env": {
        "COMFY_URL": "http://your-comfy-server-url:port",
        "OUTPUT_NODE_ID": "9",
        "PROMPT_NODE_ID": "6",
        "COMFY_WORKFLOW_JSON_FILE": "/path/to/the/comfyui_workflow_export.json"
      }
    }
  }
}

You deploy this MCP server to generate images from prompts by sending requests to a remote Comfy server, using the FastMCP framework for fast, scalable image creation with a dedicated workflow.

How to use

Start the local MCP server and then send image generation requests from your MCP client. The server runs a local process that connects to the remote Comfy server, submits prompts, and returns the generated images when ready. To begin, run the server locally and ensure your environment variables point to your Comfy setup. When you issue a request from your client, the server handles the end-to-end flow and delivers the final image.

How to install

Prerequisites you need before installing and running the server are a compatible Python runtime and the required Python packages.

# Prerequisites
Python 3.x installed

# Install the MCP CLI integration package
pip install "mcp[cli]"

Configuration and usage notes

Configure the server by setting environment variables that point to your Comfy server, the exported workflow JSON, and the specific node IDs for prompting and final output.

export COMFY_URL=http://your-comfy-server-url:port
export COMFY_WORKFLOW_JSON_FILE=/path/to/the/comfyui_workflow_export.json
export PROMPT_NODE_ID=6
export OUTPUT_NODE_ID=9

Run the server

Start the MCP server locally using Python. The server will listen for image generation requests and process them through the configured Comfy workflow.

python comfy-mcp-server.py

Available tools

generate_image

Generates an image by submitting a prompt to the configured Comfy workflow, polling for processing status, and returning the final image when ready.

Built by

VeilStrat

AI signals for GTM teams