PulseAugur
EN
LIVE 14:50:04

Comfy MCP enables AI agents to control local ComfyUI workflows

A new tool called Comfy MCP has been released, enabling coding agents to control local ComfyUI installations. This protocol allows agents to inspect models, validate workflows, and retrieve outputs. To use Comfy MCP, users need Python 3.10+, comfy-cli version 1.14.0+, and an MCP-capable client like Claude Code, Cursor, or Claude Desktop. The setup involves installing the necessary packages, configuring the ComfyUI workspace, launching ComfyUI, and then connecting the chosen MCP client. AI

IMPACT Enhances local AI model control and workflow automation for developers.

RANK_REASON This is a tool release that integrates existing AI models and software.

Read on dev.to — MCP tag →

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →

Comfy MCP enables AI agents to control local ComfyUI workflows

COVERAGE [1]

  1. dev.to — MCP tag TIER_1 English(EN) · Lachlan Dromgold ·

    Install Comfy MCP: Control Local ComfyUI from Claude Code or Cursor

    <p>Comfy MCP is Comfy's first-party local Model Context Protocol server. It lets an MCP-capable coding agent inspect the models and nodes in your ComfyUI installation, validate workflows, run them, and retrieve the outputs.</p> <p>The detail that prevents the most confusion is th…