comfy-mcp-server

comfy-mcp-server

24
cloud_platforms

Comfy MCP Server is designed to generate images based on prompts using the FastMCP framework. It connects to a remote Comfy server, submitting prompts and retrieving results. Optional integration with Ollama server allows enhanced prompt generation capabilities.

Overview

  • Project Purpose: Comfy MCP Server sets up an image generation server using the FastMCP framework to create images based on prompts through a Comfy server.
  • Key Features:
    • Interacts with a remote server to submit prompts and retrieve images.
    • Requires the uv package and project manager for Python.
    • Supports optional integration with an Ollama server for prompt generation.
  • Configuration: Environment variables must be set for the server URL, workflow file, and node IDs.
  • Usage: The server is launched using a command with configuration specified through environment variables.

Functionality

  • generate_image(prompt, ctx) generates an image by submitting a prompt to Comfy server and retrieves the processed image.
  • generate_prompt(topic, ctx) creates a detailed image generation prompt from a given topic.

Dependencies

  • Dependencies include mcp, json, urllib, time, os, langchain, and langchain-ollama.