Comfy MCP Server

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

Catalog onlyCatalog onlySTDIO

Overview

Comfy MCP Server is a server application that utilizes the FastMCP framework to generate images based on user-defined prompts by interacting with a remote Comfy server.

To use Comfy MCP Server, set the required environment variables, install the necessary packages, and launch the server using the command uvx comfy-mcp-server.

Generates images from prompts using a remote Comfy server. - Supports configuration through environment variables. - Allows for custom workflows exported from Comfy UI.

Creating images for artistic projects based on textual descriptions.
Automating image generation for design prototypes.
Integrating with applications that require dynamic image generation.

Add to your AI client

Use these steps to connect Comfy MCP Server in Cursor, Claude, VS Code, and other MCP-compatible apps. The same JSON appears in the Use with menu above for one-click copy.

Cursor

Add this to your .cursor/mcp.json file in your project root, then restart Cursor.

.cursor/mcp.json

{
  "mcpServers": {
    "comfy-mcp-server-lalanikarim": {
      "command": "npx",
      "args": [
        "-y",
        "@modelcontextprotocol/server-comfy-mcp-server-lalanikarim"
      ]
    }
  }
}

Claude Desktop

Add this server entry to the mcpServers object in your Claude Desktop config, then restart the app.

~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or %APPDATA%\Claude\claude_desktop_config.json (Windows)

{
  "mcpServers": {
    "comfy-mcp-server-lalanikarim": {
      "command": "npx",
      "args": [
        "-y",
        "@modelcontextprotocol/server-comfy-mcp-server-lalanikarim"
      ]
    }
  }
}

Claude Code

Add this to your project's .mcp.json file. Claude Code will detect it automatically.

.mcp.json (project root)

{
  "mcpServers": {
    "comfy-mcp-server-lalanikarim": {
      "command": "npx",
      "args": [
        "-y",
        "@modelcontextprotocol/server-comfy-mcp-server-lalanikarim"
      ]
    }
  }
}

VS Code (Copilot)

Add this to your .vscode/mcp.json file. Requires the GitHub Copilot extension with MCP support enabled.

.vscode/mcp.json

{
  "servers": {
    "comfy-mcp-server-lalanikarim": {
      "type": "stdio",
      "command": "npx",
      "args": [
        "-y",
        "@modelcontextprotocol/server-comfy-mcp-server-lalanikarim"
      ]
    }
  }
}

Windsurf

Add this to your Windsurf MCP config file, then restart Windsurf.

~/.codeium/windsurf/mcp_config.json

{
  "mcpServers": {
    "comfy-mcp-server-lalanikarim": {
      "command": "npx",
      "args": [
        "-y",
        "@modelcontextprotocol/server-comfy-mcp-server-lalanikarim"
      ]
    }
  }
}

Cline

Open Cline settings, navigate to MCP Servers, and add this server configuration.

Cline MCP Settings (via UI)

{
  "mcpServers": {
    "comfy-mcp-server-lalanikarim": {
      "command": "npx",
      "args": [
        "-y",
        "@modelcontextprotocol/server-comfy-mcp-server-lalanikarim"
      ]
    }
  }
}

FAQ

What is required to run Comfy MCP Server?

You need the `uv` package and a workflow file exported from Comfy UI.

Can I customize the image generation process?

Yes! You can set various environment variables to customize the workflow and output.

Is there a specific format for the prompt?

The prompt can be any string that describes the desired image.

Research and data