comfy-test

Testing infrastructure for ComfyUI custom nodes.

Test your nodes install and work correctly across Linux, macOS, Windows, and Windows Portable. No pytest code needed.

Quick Start

Add these files to your custom node repository:

1. comfy-test.toml

[test]
# Name is auto-detected from directory

[test.workflows]
cpu = "all"  # Run all workflows in workflows/ folder

2. .github/workflows/test-install.yml

name: Test Installation
on: [push, pull_request]

jobs:
  test:
    uses: PozzettiAndrea/comfy-test/.github/workflows/test-matrix.yml@main

3. workflows/test.json

A minimal ComfyUI workflow that uses your nodes. Export from ComfyUI.

Done! Push to GitHub and your tests will run automatically on all platforms.

Test Levels

comfy-test runs 7 test levels in sequence:

Level	Name	What It Does
1	SYNTAX	Check project structure (pyproject.toml/requirements.txt), CP1252 compatibility
2	INSTALL	Clone ComfyUI, create environment, install node + dependencies
3	REGISTRATION	Start server, verify nodes appear in /object_info
4	INSTANTIATION	Test each node's constructor
5	STATIC_CAPTURE	Screenshot workflows (no execution)
6	VALIDATION	4-level workflow validation (schema, graph, introspection, partial execution)
7	EXECUTION	Run workflows end-to-end, capture outputs

Each level depends on previous levels. You can run up to a specific level with --level:

comfy-test run --level registration  # Runs: SYNTAX -> INSTALL -> REGISTRATION

Workflow Validation (4 Levels)

The VALIDATION level runs comprehensive checks before execution:

Level	Name	What It Checks
1	Schema	Widget values match allowed enums, types, and ranges
2	Graph	Connections are valid, all referenced nodes exist
3	Introspection	Node definitions are well-formed (INPUT_TYPES, RETURN_TYPES, FUNCTION)
4	Partial Execution	Runs non-CUDA nodes to verify they work

Detecting CUDA Nodes

To mark nodes as requiring CUDA (excluded from partial execution), use comfy-env.toml:

[cuda]
packages = ["nvdiffrast", "flash-attn"]

Configuration Reference

Minimal Config

[test]
# Everything has sensible defaults - this is all you need

[test.workflows]
cpu = "all"

Full Config Example

[test]
# Name is auto-detected from directory name (e.g., "ComfyUI-MyNode")

# ComfyUI version to test against
comfyui_version = "latest"  # or a tag like "v0.2.0" or commit hash

# Python version (default: random from 3.11, 3.12, 3.13)
python_version = "3.11"

# Test levels to run (default: all)
# Options: syntax, install, registration, instantiation, static_capture, validation, execution
levels = ["syntax", "install", "registration", "instantiation", "static_capture", "validation", "execution"]
# Or use: levels = "all"

# Enable/disable platforms (all enabled by default)
[test.platforms]
linux = true
macos = true
windows = true
windows_portable = true

# Workflow configuration
[test.workflows]
# Workflows to run on CPU runners (GitHub-hosted)
cpu = "all"  # or list specific files: ["test_basic.json", "test_advanced.json"]

# Workflows to run on GPU runners (self-hosted)
gpu = ["test_gpu.json"]

# Timeout for workflow execution in seconds (default: 3600)
timeout = 120

# Platform-specific settings
[test.linux]
enabled = true
skip_workflow = false  # Skip workflow execution, only verify registration

[test.macos]
enabled = true
skip_workflow = false

[test.windows]
enabled = true
skip_workflow = false

[test.windows_portable]
enabled = true
skip_workflow = false
comfyui_portable_version = "latest"  # Portable-specific version

Workflow Discovery

Workflows are auto-discovered from the workflows/ folder:

• All .json files in workflows/ are found automatically
• Use cpu = "all" to run all discovered workflows on CPU
• Use gpu = "all" to run all discovered workflows on GPU
• Or specify individual files: cpu = ["basic.json", "advanced.json"]

CLI

# Install
pip install comfy-test

# Initialize config and GitHub workflow
comfy-test init

# Run tests locally
comfy-test run --platform linux

# Run specific level only
comfy-test run --level registration

# Publish results to GitHub Pages
comfy-test publish ./results --repo owner/repo

CUDA Packages on CPU-only CI

comfy-test runs on CPU-only GitHub Actions runners. For nodes that use CUDA packages:

1. Installation works - comfy-test sets COMFY_ENV_CUDA_VERSION=12.8 so comfy-env can resolve wheel URLs
2. Import may fail - CUDA packages typically fail to import without a GPU

For full CUDA testing, use a self-hosted runner with a GPU.

License

MIT