📌Overview

Purpose: The Comfy MCP Server is designed to generate images based on prompts by leveraging a remote Comfy server through the FastMCP framework.

Overview: This server setup interacts with a Comfy server to facilitate image generation. It allows users to submit prompts via a defined workflow and receive the resulting images efficiently.

Key Features:

Remote Interaction: Connects seamlessly with a remote Comfy server to process image generation prompts.
Custom Workflow Support: Requires user-defined workflow files from Comfy UI, allowing for tailored image generation processes based on specific needs.
Flexible Output Options: Users can choose to receive generated images as URLs or file downloads, enhancing versatility for different applications.
Environment Configuration: The server supports various configurable environment variables, ensuring flexibility in connecting to different Comfy servers and workflows.

Comfy MCP Server

A server using FastMCP framework to generate images based on prompts via a remote Comfy server.

Overview

This script sets up a server using the FastMCP framework to generate images based on prompts using a specified workflow. It interacts with a remote Comfy server to submit prompts and retrieve generated images.

Prerequisites

uv package and project manager for Python.
Workflow file exported from Comfy UI. This code includes a sample Flux-Dev-ComfyUI-Workflow.json for reference. You need to export your own workflow and set the environment variables accordingly.

Install the required packages for local development:

uvx mcp[cli]

Configuration

Set the following environment variables:

COMFY_URL: URL of your Comfy server.
COMFY_WORKFLOW_JSON_FILE: Absolute path to the API export JSON file for the ComfyUI workflow.
PROMPT_NODE_ID: ID of the text prompt node.
OUTPUT_NODE_ID: ID of the output node with the final image.
OUTPUT_MODE: url or file to select desired output format.

Optionally, connect to an Ollama server for prompt generation:

OLLAMA_API_BASE: URL where Ollama is running.
PROMPT_LLM: Name of the model hosted on Ollama for prompt generation.

Example:

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
export OUTPUT_MODE=file

Usage

Launch Comfy MCP Server with:

uvx comfy-mcp-server

Example Claude Desktop Config

{
  "mcpServers": {
    "Comfy MCP Server": {
      "command": "/path/to/uvx",
      "args": [
        "comfy-mcp-server"
      ],
      "env": {
        "COMFY_URL": "http://your-comfy-server-url:port",
        "COMFY_WORKFLOW_JSON_FILE": "/path/to/the/comfyui_workflow_export.json",
        "PROMPT_NODE_ID": "6",
        "OUTPUT_NODE_ID": "9",
        "OUTPUT_MODE": "file"
      }
    }
  }
}

Functionality

`generate_image(prompt: str, ctx: Context) -> Image | str`

Generates an image from a prompt:

Checks if required environment variables are set.
Loads prompt template from a JSON file.
Submits the prompt to the Comfy server.
Polls the server for processing status.
Retrieves and returns the generated image.

`generate_prompt(topic: str, ctx: Context) -> str`

Generates a detailed image generation prompt from a specified topic.

Dependencies

mcp: FastMCP server framework.
json: JSON data handling.
urllib: HTTP requests.
time: Delays for polling.
os: Environment variable access.
langchain: LLM prompt chain creation.
langchain-ollama: Ollama modules for LangChain.

License

This project is licensed under the MIT License - see the LICENSE file for details.