ComfyUI MCP

A lightweight MCP server for a local ComfyUI instance that lets AI agents generate and iteratively refine images, audio, and video. Agents can submit workflows, override prompts and sampler parameters, poll job status, and retrieve outputs, which makes conversational iteration on a local diffusion pipeline practical.

{
  "mcpServers": {
    "comfyui": {
      "command": "npx",
      "args": [],
      "env": {
        "COMFYUI_URL": "<YOUR_COMFYUI_URL>"
      }
    }
  }
}

{
  "type": "object",
  "required": [
    "prompt"
  ],
  "properties": {
    "prompt": {
      "type": "string",
      "description": "Positive prompt text"
    },
    "workflow": {
      "type": "string",
      "description": "Name of the workflow to run"
    },
    "negative_prompt": {
      "type": "string",
      "description": "Negative prompt text"
    }
  }
}

{
  "type": "object",
  "required": [
    "prompt_id"
  ],
  "properties": {
    "prompt_id": {
      "type": "string",
      "description": "Job identifier returned when the workflow was queued"
    }
  }
}