How do I use ComfyUI MCP Server?

1. Click on "Deploy Server". 2. Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state. 3. In the chat, type @ followed by the MCP server name and your instructions, e.g., "@ComfyUI MCP Server execute this workflow: checkpoint: CheckpointLoaderSimple ckpt_name: sd_xl_base_1.0.safetensors" That's it! The server will respond to your query, and you can continue using it as needed. Here is a step-by-step guide with screenshots.

ComfyUI MCP Server

by christian-byrne

Overview Schema Source Related Servers Discussions Admin

Python

Hybrid

ComfyUI MCP Server

DSL-first workflow management for ComfyUI via Model Context Protocol (MCP)

A production-ready MCP server that enables AI agents to manage ComfyUI workflows using a human-readable Domain Specific Language (DSL). The core design philosophy is DSL-first: agents work entirely in DSL format, with JSON conversion happening transparently.

🚀 Quick Start

Installation

pip install comfy-mcp

Usage with Claude Code

Create MCP configuration:

{
  "mcpServers": {
    "comfyui-workflows": {
      "command": "comfy-mcp",
      "args": [],
      "env": {}
    }
  }
}

Start Claude Code with MCP:

claude --mcp-config mcp_config.json

Use in conversation:

"Execute this workflow: [paste DSL]"
"List workflows in examples directory"
"Show ComfyUI queue status"

Related MCP server: ComfyUI MCP Server

✨ Features

🔄 DSL-First Design

Agents work entirely in human-readable DSL
Automatic JSON ↔ DSL conversion
No need to think about format conversion

📁 File Operations

read_workflow - Auto-converts JSON to DSL
write_workflow - Saves DSL as JSON/DSL
list_workflows - Discovers workflow files
validate_workflow - DSL syntax validation
get_workflow_info - Workflow analysis

⚡ Execution Operations

execute_workflow - Run DSL workflows on ComfyUI
get_job_status - Monitor execution & download images
list_comfyui_queue - View ComfyUI queue status

🎨 DSL Syntax Example

## Model Loading

checkpoint: CheckpointLoaderSimple
  ckpt_name: sd_xl_base_1.0.safetensors

## Text Conditioning

positive: CLIPTextEncode
  text: a beautiful landscape, detailed, photorealistic
  clip: @checkpoint.clip

negative: CLIPTextEncode
  text: blurry, low quality
  clip: @checkpoint.clip

## Generation

latent: EmptyLatentImage
  width: 1024
  height: 1024

sampler: KSampler
  model: @checkpoint.model
  positive: @positive.conditioning
  negative: @negative.conditioning
  latent_image: @latent.latent
  seed: 42
  steps: 20

## Output

decode: VAEDecode
  samples: @sampler.latent
  vae: @checkpoint.vae

save: SaveImage
  images: @decode.image
  filename_prefix: output

🏗️ Architecture

┌─────────────────┐    ┌──────────────┐    ┌─────────────┐
│   AI Agent      │────│  MCP Server  │────│  ComfyUI    │
│   (Claude)      │    │              │    │   Server    │
└─────────────────┘    └──────────────┘    └─────────────┘
         │                       │                  │
         │ DSL Workflows         │ JSON API         │
         │                       │                  │
         ▼                       ▼                  ▼
   Natural Language ────► DSL Parser ────► JSON Converter

Key Components:

DSL Parser: Converts human-readable DSL to Abstract Syntax Tree
JSON Converter: Bidirectional conversion between DSL and ComfyUI JSON
MCP Server: Exposes tools via Model Context Protocol
Execution Engine: Integrates with ComfyUI API for workflow execution

📖 Documentation

Core Classes

DSLParser: Parse DSL text into Abstract Syntax Tree
DslToJsonConverter: Convert DSL AST to ComfyUI JSON
JsonToDslConverter: Convert ComfyUI JSON to DSL AST

MCP Tools

Tool	Description	Example
`read_workflow`	Read and convert workflows to DSL	`read_workflow("workflow.json")`
`write_workflow`	Write DSL to disk as JSON/DSL	`write_workflow("output.json", dsl)`
`list_workflows`	Find workflow files	`list_workflows("./workflows")`
`validate_workflow`	Check DSL syntax	`validate_workflow(dsl_content)`
`get_workflow_info`	Analyze structure	`get_workflow_info(dsl_content)`
`execute_workflow`	Run on ComfyUI	`execute_workflow(dsl_content)`
`get_job_status`	Monitor execution	`get_job_status(prompt_id)`
`list_comfyui_queue`	View queue	`list_comfyui_queue()`

🛠️ Development

Setup

git clone https://github.com/christian-byrne/comfy-mcp.git
cd comfy-mcp
pip install -e ".[dev]"
pre-commit install

Testing

# Run all tests
pytest

# Run with coverage
pytest --cov=comfy_mcp --cov-report=html

# Run specific test types
pytest -m unit
pytest -m integration
pytest -m "not slow"

Code Quality

# Format code
black .

# Lint code  
ruff check .

# Type checking
mypy comfy_mcp

Documentation

cd docs
make html

🔧 Configuration

Environment Variables

COMFYUI_SERVER: ComfyUI server address (default: 127.0.0.1:8188)
MCP_DEBUG: Enable debug logging
MCP_LOG_LEVEL: Set log level (DEBUG, INFO, WARNING, ERROR)

ComfyUI Setup

Install ComfyUI
Start server: python main.py --listen 0.0.0.0
Ensure models are installed in models/checkpoints/

🤝 Contributing

We welcome contributions! Please see CONTRIBUTING.md for guidelines.

Development Workflow

Fork the repository
Create a feature branch: git checkout -b feature-name
Make changes and add tests
Run tests and linting: pytest && black . && ruff check .
Submit a pull request

📄 License

This project is licensed under the MIT License - see the LICENSE file for details.

🙏 Acknowledgments

ComfyUI - Amazing stable diffusion GUI
FastMCP - Excellent MCP framework
Anthropic - Model Context Protocol specification

📈 Roadmap

v0.2.0: Enhanced DSL features (templates, macros)
v0.3.0: Web UI for workflow management
v0.4.0: Git integration for workflow versioning
v0.5.0: ComfyUI node discovery and documentation
v1.0.0: Production deployment features

Built with ❤️ for the ComfyUI and AI automation community

Available Tools

16 tools

execute_workflowA

Execute a DSL workflow on ComfyUI server.

Converts DSL to JSON and submits to ComfyUI for execution. Can optionally wait for completion and return results.

Args: dsl: Workflow content in DSL format server_address: ComfyUI server address (default: 127.0.0.1:8188) wait_for_completion: Whether to wait for execution to complete timeout_seconds: Maximum time to wait for completion

Returns: Execution result with prompt_id, status, and outputs if completed

Examples: execute_workflow(dsl_content) execute_workflow(dsl_content, server_address="192.168.1.100:8188") execute_workflow(dsl_content, wait_for_completion=False)

Inputs: dsl (required), server_address, wait_for_completion, timeout_seconds. Returns: object

generate_from_templateA

Generate a workflow from a template with custom parameters.

Creates a complete DSL workflow by substituting parameters into the template. Optionally saves to file.

Args: template_name: Name of the template to use parameters: Dictionary of parameter values to substitute save_path: Optional path to save the generated workflow

Returns: Generated DSL content and validation results

Examples: generate_from_template("text2img_basic", {"prompt": "sunset"}) generate_from_template("img2img", {"image_path": "input.png"}, "workflows/my_img2img.dsl")

Inputs: template_name (required), parameters, save_path. Returns: object

get_job_statusA

Get status and results of a ComfyUI job.

Checks execution status and optionally downloads generated images.

Args: prompt_id: The prompt ID returned by execute_workflow server_address: ComfyUI server address download_images: Whether to download generated images image_save_path: Directory to save images (relative to workflows/)

Returns: Job status with completion info and image paths if downloaded

Examples: get_job_status("12345-abcde-67890") get_job_status("12345-abcde-67890", download_images=True)

Inputs: prompt_id (required), server_address, download_images, image_save_path. Returns: object

get_templateA

Get detailed information about a specific template.

Retrieves complete template information including parameters, requirements, and DSL preview.

Args: template_name: Name of the template to retrieve

Returns: Template details with parameters and DSL preview

Examples: get_template("text2img_basic") get_template("controlnet_pose")

Inputs: template_name (required). Returns: object

get_template_dslA

Get DSL content for any template (custom or official).

Retrieves the DSL representation of a template, which can then be modified or executed directly.

Args: template_name: Name of the template source: Template source ("custom", "official", or "auto")

Returns: Template DSL content and metadata

Examples: get_template_dsl("text2img_basic") get_template_dsl("openai_dalle_3_text_to_image", "official")

Inputs: template_name (required), source. Returns: object

get_workflow_infoB

Analyze workflow structure and return metadata.

Parses DSL and extracts structural information like node types, sections, and connections without executing the workflow.

Args: dsl: Workflow content in DSL format

Returns: Workflow metadata including nodes, sections, and connections

Examples: get_workflow_info(dsl_content)

Inputs: dsl (required). Returns: object

list_comfyui_queueA

List current ComfyUI execution queue.

Shows running and pending jobs in the ComfyUI queue.

Args: server_address: ComfyUI server address

Returns: Queue information with running and pending jobs

Examples: list_comfyui_queue() list_comfyui_queue("192.168.1.100:8188")

Inputs: server_address. Returns: object

list_official_templatesB

List official ComfyUI templates.

Returns templates from the official Comfy-Org repository that have been synced and converted to DSL format.

Returns: List of official template metadata

Examples: list_official_templates()

Inputs: no named parameters. Returns: result

list_templatesA

List available workflow templates.

Discover workflow templates by category, difficulty, or tags. Templates provide ready-to-use workflows for common use cases.

Args: category: Filter by category (Generation, Enhancement, Editing, etc.) difficulty: Filter by difficulty (beginner, intermediate, advanced) tag: Filter by tag (text2img, inpainting, controlnet, etc.)

Returns: List of template metadata with names, descriptions, and parameters

Examples: list_templates() list_templates(category="Generation") list_templates(difficulty="beginner") list_templates(tag="text2img")

Inputs: category, difficulty, tag. Returns: result

list_workflowsA

List workflow files in a directory.

Discovers workflow files (.json and .dsl) in the specified directory. Supports glob patterns for filtering.

Args: directory: Directory to search (default: "workflows") pattern: Glob pattern for filtering (default: "*" for all files)

Returns: List of workflow info dicts with name, size, modified time

Examples: list_workflows() list_workflows("workflows", "*.json") list_workflows("../dsl/examples/dsl")

Inputs: directory, pattern. Returns: result

read_workflowA

Read a workflow file and return it as DSL format.

Supports both JSON and DSL input files. Automatically detects format and converts JSON to DSL transparently.

Args: filepath: Path to workflow file (.json or .dsl)

Returns: Workflow content in DSL format

Examples: read_workflow("workflows/my_workflow.json") read_workflow("../dsl/examples/dsl/simple.dsl")

Inputs: filepath (required). Returns: result

search_templatesB

Search templates by name, description, or tags.

Performs fuzzy search across template metadata to find relevant workflow templates.

Args: query: Search query (searches name, description, tags)

Returns: List of matching templates

Examples: search_templates("pose") search_templates("upscaling") search_templates("text to image")

Inputs: query (required). Returns: result

sync_official_templatesA

Sync official ComfyUI templates from GitHub.

Downloads and processes official workflow templates from the Comfy-Org/workflow_templates repository. Converts them to DSL format for use with the template system.

Returns: Sync status with count of successfully processed templates

Examples: sync_official_templates()

Inputs: no named parameters. Returns: object

validate_template_parametersA

Validate parameters for a template without generating.

Checks if provided parameters are valid for the template and returns detailed validation results.

Args: template_name: Name of the template parameters: Dictionary of parameters to validate

Returns: Validation results with errors and warnings

Examples: validate_template_parameters("text2img_basic", {"width": "512"})

Inputs: template_name (required), parameters (required). Returns: object

validate_workflowA

Validate DSL workflow syntax.

Parses DSL and checks for syntax errors without executing the workflow. Returns validation status and any errors found.

Args: dsl: Workflow content in DSL format

Returns: Validation result with is_valid, errors, and warnings

Examples: validate_workflow(dsl_content)

Inputs: dsl (required). Returns: object

write_workflowA

Write a workflow to disk.

Takes DSL content and writes it to disk. By default, converts to JSON format. Can optionally save as .dsl format directly.

Args: filepath: Destination file path dsl: Workflow content in DSL format format: Output format ("json" or "dsl", default: "json")

Returns: Status dict with path, size, and format info

Examples: write_workflow("workflows/new_workflow.json", dsl_content) write_workflow("workflows/backup.dsl", dsl_content, format="dsl")

Inputs: filepath (required), dsl (required), format. Returns: object

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

Deploy Server

Claude Code, inside the Godot 4 editorFlockbay puts an AI agent in the real Godot 4 editor: it writes GDScript, runs your game and fixes what breaks. Free, or bring your own Claude Code or Codex.flockbay.comAd

TDQS

A3.9/5.0

Scored 2026-04-18 08:20 across 16 tools

Disambiguation4/5

Most tools have distinct purposes, but there is some overlap between get_template and get_template_dsl, as both retrieve template details, which could cause confusion. However, descriptions clarify that get_template provides metadata while get_template_dsl focuses on DSL content, helping to differentiate them.

Naming Consistency5/5

All tool names follow a consistent verb_noun pattern with snake_case, such as execute_workflow, list_templates, and validate_workflow. This predictability makes the set easy to navigate and understand.

Tool Count4/5

With 16 tools, the count is slightly high but reasonable for a ComfyUI server covering workflow execution, template management, and file operations. It feels comprehensive without being overwhelming, though some tools like get_template and get_template_dsl could potentially be consolidated.

Completeness5/5

The tool set provides complete coverage for the domain, including workflow execution (execute_workflow), status tracking (get_job_status), template handling (list_templates, generate_from_template), file operations (read_workflow, write_workflow), and validation (validate_workflow). There are no obvious gaps for core ComfyUI operations.

Maintenance

ActivityInactive

ResponsivenessNo recent community issues

Resources

Related MCP Connectors

studio
io.kaitoi
Build and run visual creative-production workflows from your AI agent.
AI & Machine Learning Design Tools App Automation
2
https://mcp.studio.kaitoi.io
neonloopsA
com.neonloops
Build, test and publish agentic workflows from your AI client: approve steps, read tables, see runs.
Agent Orchestration Autonomous Agents Developer Tools
151
https://app.neonloops.com/api/mcp/v1
runcomfy-mcpA
io.github.runcomfy-com
Remote MCP for RunComfy: ComfyUI deployments, hosted models, LoRA training. 31 tools.
AI & Machine Learning Cloud Platforms Image & Video Processing
2
https://mcp.runcomfy.com/mcp
OutcomeCI
io.github.outcomeci
Create, validate, and manage agent-powered workflows with OutcomeCI.
Autonomous Agents Agent Orchestration App Automation
11
https://api.outcomeci.com/v1/mcp

Related MCP Servers

ComfyUI MCP Server
Image & Video Processing App Automation Developer Tools
neutrinotek
F
license
B
quality
D
maintenance
Enables orchestration of ComfyUI workflows through natural language by discovering workflow templates, applying mutations, and submitting prompts to running ComfyUI instances. Provides asset validation and lightweight tooling for AI image generation experimentation.
Updated a year ago (2025-10-29 02:57 UTC)
3
-
ComfyUI MCP Server
Image & Video Processing App Automation Developer Tools
IO-AtelierTech
A
license
A
quality
D
maintenance
Enables comprehensive ComfyUI workflow automation including image generation, workflow management, node discovery, and system monitoring through natural language interactions with local or remote ComfyUI servers.
Updated a year ago (2025-12-28 06:41 UTC)
31
52 PyPI
14
MIT
ComfyUI MCP Server
Image & Video Processing Multimedia Processing Audio Processing
jaewilson07
A
license
Not graded
quality
Not graded
maintenance
Enables AI agents to generate and iteratively refine images, audio, and video by interacting with a local ComfyUI instance through natural conversation. It provides comprehensive tools for workflow management, node introspection, and publishing generated assets.
Updated 9 months ago (2026-02-12 20:05 UTC)
Apache 2.0
ComfyUI MCP Hub
AI & Machine Learning Image & Video Processing
ArboRithmDev
A
license
Not graded
quality
D
maintenance
Give AI agents full control over your local ComfyUI by exposing 77 tools for workflow management, image generation, model handling, and real-time canvas control.
Updated 7 months ago (2026-04-09 07:32 UTC)
4
AGPL 3.0

ComfyUI MCP Server

🚀 Quick Start

Installation

Usage with Claude Code

✨ Features

🔄 DSL-First Design

📁 File Operations

⚡ Execution Operations

🎨 DSL Syntax Example

🏗️ Architecture

📖 Documentation

Core Classes

MCP Tools

🛠️ Development

Setup

Testing

Code Quality

Documentation

🔧 Configuration

Environment Variables

ComfyUI Setup

🤝 Contributing

Development Workflow

📄 License

🙏 Acknowledgments

📈 Roadmap

Available Tools

Tool Schema Changelog

TDQS

Maintenance

Resources

Related MCP Connectors

Related MCP Servers

ComfyUI MCP Server

ComfyUI MCP Server

ComfyUI MCP Server

ComfyUI MCP Hub