القائمة
Use this skill to generate AI images using ModelScope's Tongyi-MAI/Z-Image-Turbo model. Simply describe the image you want and it will be generated. Supports Chinese and English prompts.
التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.
استنادا إلى تصنيف SOC المهني
| name | modelscope |
| description | Use this skill to generate AI images using ModelScope's Tongyi-MAI/Z-Image-Turbo model. Simply describe the image you want and it will be generated. Supports Chinese and English prompts. |
Generate AI images using ModelScope's powerful text-to-image model.
Trigger when user:
# Generate an image with a text prompt
python scripts/run.py generate.py --prompt "A golden cat sitting on a windowsill"
# Generate with Chinese prompt
python scripts/run.py generate.py --prompt "一只金色的猫坐在窗台上，阳光照进来"
# Generate and save to specific path
python scripts/run.py generate.py --prompt "A beautiful sunset over mountains" --save-path "sunset.png"
# Generate with custom model
python scripts/run.py generate.py --prompt "Cyberpunk city at night" --model "Tongyi-MAI/Z-Image"
Create a .env file in the skill directory:
MODELSCOPE_API_KEY=ms-your-api-key-here
Or set as environment variable:
# Windows PowerShell
$env:MODELSCOPE_API_KEY="ms-your-api-key-here"
# Windows CMD
set MODELSCOPE_API_KEY=ms-your-api-key-here
# Linux/Mac
export MODELSCOPE_API_KEY="ms-your-api-key-here"
python scripts/run.py generate.py --help
This will automatically:
.venv virtual environment| Parameter | Type | Required | Default | Description |
|---|---|---|---|---|
--prompt | string | Yes | - | Text description of the image to generate |
--model | string | No | Tongyi-MAI/Z-Image-Turbo | Model ID to use |
--save-path | string | No | result_image.jpg | File path to save the image |
--timeout | integer | No | 300 | Maximum wait time in seconds |
--async-mode | boolean | No | true | Use async mode for generation |
# Simple image generation
python scripts/run.py generate.py --prompt "A golden cat"
# With custom save location
python scripts/run.py generate.py --prompt "A majestic dragon" --save-path "dragon.png"
# With longer timeout for complex images
python scripts/run.py generate.py --prompt "A detailed fantasy landscape with castles and dragons" --timeout 600
# Using the standard quality model
python scripts/run.py generate.py --prompt "A photorealistic portrait" --model "Tongyi-MAI/Z-Image"
| Model ID | Description | Speed | Quality |
|---|---|---|---|
| Tongyi-MAI/Z-Image-Turbo | Fast generation | Fast | Good |
| Tongyi-MAI/Z-Image | Standard generation | Medium | High |
| Problem | Solution |
|---|---|
MODELSCOPE_API_KEY not set | Set your API key in .env file or environment variable |
Image generation failed | Check your API key is valid and prompt follows content policies |
Connection timeout | Increase --timeout parameter (default: 300 seconds) |
ModuleNotFoundError | Dependencies will auto-install on first run, or manually run pip install -r requirements.txt |
Be Descriptive: More detailed prompts produce better images
"A golden cat" → "A fluffy golden cat with emerald eyes sitting on a velvet cushion"
Specify Style: Include art style for better control
"A mountain landscape in the style of Studio Ghibli"
Use Chinese or English: The model supports both languages well
"一只在月光下的猫，水彩画风格"
Add Details: Include lighting, mood, and composition
"A serene sunset over calm ocean waters, warm orange and pink sky, peaceful atmosphere"
~/.claude/skills/modelscope-image-gen/
├── SKILL.md # This file - skill documentation
├── scripts/
│ ├── run.py # Wrapper script for environment management
│ └── generate.py # Main image generation script
├── data/ # Local storage (auto-created)
├── .env # API key configuration (create this)
├── requirements.txt # Python dependencies
└── .venv/ # Virtual environment (auto-created)