Skip to main content
MCP server CX Codex

comfy-mcp — Claude Code MCP server: what it does, how to install

comfy-mcp is a Claude Code MCP server that drives a local ComfyUI instance to generate images and monitor jobs.

no license ★ 366 updated 2026-11-17

Install into Claude Code · Platform CLI All methods and platforms ↓
claude mcp add comfy-mcp -- uvx comfy-mcp
Open on GitHub

What comfy-mcp does

The server connects AI agents to a local ComfyUI installation via the comfy-cli engine. It executes workflow files, polls job status, inspects outputs, and introspects live nodes and models.

Python 3.10 or newer and a working comfy binary in PATH are required before starting the server.

What it helps with

  • generate images from text prompts
  • monitor and cancel background rendering jobs
  • validate and build workflow graphs
  • manage the ComfyUI server process

Example prompts for the agent

  • Generate an image using the local ComfyUI workflow
  • Check the status of the current render job
  • Restart the local ComfyUI server

How to install comfy-mcp

Platform CLI

claude mcp add comfy-mcp -- uvx comfy-mcp

uvx runs the command named after the package; if the README names another one, use uvx --from comfy-mcp <command>

Step by step, with commands for each platform: How to add an MCP server to Claude Code, Codex, Cursor, VS Code, Gemini CLI and more

Questions about comfy-mcp

How to install comfy-mcp in Claude Code?

Run claude mcp add comfy-mcp -- uvx comfy-mcp.

What does comfy-mcp do?

It is an MCP server that lets AI agents control local ComfyUI instances, run workflows, and inspect generated media.

What do I need to run comfy-mcp?

You need Python 3.10 or later and the comfy-cli package installed on your system.

Similar MCP servers

All MCP servers for Agent setup: memory, context, orchestration →

Project sponsors

Companies that keep this analytics open