image-generator

star 0

AI image generation with support for multiple providers and artistic styles. Use when users request image generation, creation, or visualization tasks such as "generate an image of...", "create a picture showing...", "make an illustration of...", or when they specify artistic styles like anime, realistic, cartoon, oil painting, etc.

STDSuperman By STDSuperman schedule Updated 1/27/2026

name: image-generator description: AI image generation with support for multiple providers and artistic styles. Use when users request image generation, creation, or visualization tasks such as "generate an image of...", "create a picture showing...", "make an illustration of...", or when they specify artistic styles like anime, realistic, cartoon, oil painting, etc.

Image Generator

Generate AI images using various providers and artistic styles.

Quick Start

Generate an image with a simple prompt:

python scripts/generate_image.py "A golden cat sitting on a cloud"

Generate with a specific style:

python scripts/generate_image.py "A golden cat" --style anime

Configuration

  1. Copy .env.example to .env in the skill directory
  2. Add your API keys:
    MODELSCOPE_TOKEN=your_token_here
    JIEKOU_API_KEY=your_jiekou_key_here
    JIEKOU_ENABLE=true
    

Supported Providers

Currently supported:

  • ModelScope (default): Tongyi-MAI/Z-Image-Turbo model
  • Jiekou.ai (即梦 4.5): seedream-4.5 model

For adding new providers, see references/providers.md.

Style Templates

The skill includes built-in style templates to enhance prompts. Available styles:

  • realistic - Photorealistic images
  • anime - Japanese animation style
  • cartoon - Western cartoon style
  • oil-painting - Classical oil painting
  • watercolor - Watercolor painting
  • sketch - Pencil sketch
  • 3d-render - 3D rendered style
  • cyberpunk - Cyberpunk aesthetic
  • fantasy - Fantasy art style

For complete style descriptions and customization, see references/styles.md.

Script Usage

python scripts/generate_image.py <prompt> [options]

Options:
  --style STYLE         Apply a style template (see references/styles.md)
  --provider PROVIDER   Choose provider (default: modelscope)
  --output PATH         Output file path (default: generated_image.png)
  --model MODEL         Specific model ID (provider-dependent)

Workflow

  1. Load API credentials from .env
  2. Apply style template if specified
  3. Call the selected provider's API
  4. Poll for completion (async providers)
  5. Save the generated image

Examples

Generate a realistic portrait:

python scripts/generate_image.py "A woman with blue eyes" --style realistic

Generate anime-style artwork:

python scripts/generate_image.py "A magical girl" --style anime --output magical_girl.png
Install via CLI
npx skills add https://github.com/STDSuperman/skills --skill image-generator
Repository Details
star Stars 0
call_split Forks 0
navigation Branch main
article Path SKILL.md
More from Creator