Gemini Image Generator

Generate high-quality images for projects using Google's Gemini 3 Pro Image Preview model. This skill provides workflows for creating various types of project images including icons, illustrations, banners, and concept art.

Overview

The Gemini 3 Pro Image Preview model (gemini-3-pro-image-preview) offers native image generation capabilities through the Generative Language API. It supports:

• Text-to-image generation
• Image editing and transformation
• Style transfer
• Multi-image composition

Prerequisites

Before using this skill:

1. Obtain a Gemini API key from Google AI Studio
2. Set the environment variable: export GEMINI_API_KEY="your-api-key"

Quick Start

Generate Image via Python Script

Execute the bundled script to generate images:

python3 "${SKILL_DIR}/scripts/generate_image.py" \
  --prompt "A cute banana character mascot for a mobile app, kawaii style, yellow and brown colors" \
  --output "./generated_image.png"

Generate Image via cURL

For direct API calls without dependencies:

curl -s -X POST \
  "https://generativelanguage.googleapis.com/v1beta/models/gemini-3-pro-image-preview:generateContent?key=${GEMINI_API_KEY}" \
  -H "Content-Type: application/json" \
  -d '{
    "contents": [{
      "parts": [{"text": "Your prompt here"}]
    }],
    "generationConfig": {
      "responseModalities": ["TEXT", "IMAGE"]
    }
  }' | python3 -c "
import sys, json, base64
data = json.load(sys.stdin)
for part in data.get('candidates', [{}])[0].get('content', {}).get('parts', []):
    if 'inlineData' in part:
        img_data = base64.b64decode(part['inlineData']['data'])
        with open('output.png', 'wb') as f:
            f.write(img_data)
        print('Image saved to output.png')
"

Image Generation Workflows

Workflow 1: Project Icon Generation

Generate app icons or project logos:

python3 "${SKILL_DIR}/scripts/generate_image.py" \
  --prompt "Modern flat design app icon for [PROJECT_TYPE], minimalist style, vibrant colors, suitable for iOS/Android" \
  --output "./icon.png" \
  --aspect-ratio "1:1"

Workflow 2: Banner/Hero Image

Create marketing banners or hero images:

python3 "${SKILL_DIR}/scripts/generate_image.py" \
  --prompt "Professional banner image for [PROJECT_NAME], modern tech aesthetic, gradient background" \
  --output "./banner.png" \
  --aspect-ratio "16:9"

Workflow 3: Illustration Generation

Generate illustrations for documentation or UI:

python3 "${SKILL_DIR}/scripts/generate_image.py" \
  --prompt "Clean vector-style illustration showing [CONCEPT], soft colors, professional look" \
  --output "./illustration.png"

Workflow 4: Image Editing/Transformation

Transform or edit existing images:

python3 "${SKILL_DIR}/scripts/generate_image.py" \
  --prompt "Transform this image into a watercolor painting style while preserving the main subject" \
  --input "./source_image.png" \
  --output "./transformed.png"

Prompt Engineering Tips

Effective Prompt Structure

[Subject] + [Style] + [Details] + [Technical specs]

Example:

A friendly robot mascot (subject)
in pixel art style (style)
with blue and orange colors, waving hand (details)
on transparent background, 512x512 resolution (technical)

Style Keywords

Project-Specific Prompts

For different project types:

• Mobile App: "Modern app icon, rounded corners, gradient background, single symbolic element"
• Web Dashboard: "Clean UI illustration, data visualization theme, blue corporate colors"
• Game: "Game asset sprite, detailed pixel art, fantasy theme, transparent background"
• Documentation: "Technical diagram style, clean vector illustration, explanatory visual"

Configuration Options

Aspect Ratios

Image Sizes

Error Handling

Common issues and solutions:

Bundled Resources

Scripts

• scripts/generate_image.py - Main image generation script with full configuration options
• scripts/batch_generate.py - Generate multiple images from a prompt list

References

• references/api-reference.md - Complete Gemini API documentation
• references/prompt-templates.md - Ready-to-use prompt templates for various project types

Examples

• examples/generate_icon.sh - Example: Generate app icon
• examples/generate_banner.sh - Example: Generate project banner
• examples/batch_config.json - Example: Batch generation configuration

Best Practices

1. Be Specific: Include detailed descriptions of desired output
2. Specify Style: Always mention the artistic style explicitly
3. Define Colors: List specific colors when brand consistency matters
4. Set Constraints: Specify aspect ratio and size requirements upfront
5. Iterate: Generate multiple variations and refine prompts based on results
6. Save Prompts: Document successful prompts for future consistency

Integration Notes

When integrating generated images into projects:

1. Check image dimensions match target requirements
2. Verify file format compatibility (PNG recommended for transparency)
3. Consider compression for web assets
4. Store original prompts alongside images for reproducibility