image-generation: AI Image Generation for Claude Code Developers

AI image generation has become an essential skill for modern developers, especially those building creative applications like short video makers, educational content platforms, and interactive storytelling tools. The image-generation skill pack brings two powerful, production-ready image generation capabilities to your Claude Code environment — one for creating stunning static images and another for transforming those images into engaging short video clips.

Gemini NanoBanana: High-Quality Static Image Generation

The image-nanobanana skill leverages Google's Gemini NanoBanana model to generate high-quality AI images with specific artistic styles. Unlike generic image generators, this implementation is optimized for practical development use cases:

• Automatic style prefixes — Apply Ghibli, Educational, or other artistic styles without complex prompt engineering
• Portrait-optimized — Default 9:16 aspect ratio perfect for YouTube Shorts, TikTok, and Instagram Reels
• Character reference mode — Maintain consistent character appearance across multiple images
• Production-ready TypeScript interface — Clean, well-documented API for easy integration

The implementation uses the official @google/generative-ai SDK and follows best practices for cost optimization. Each image generation call incurs API costs, so the skill includes built-in safeguards to minimize unnecessary calls while maintaining quality.

Seedance 1.5: Professional Image-to-Video Conversion

Once you have your perfect static images, the image-seedance skill transforms them into captivating short video clips. This implementation uses Seedance 1.5 Fast model via Atlas Cloud API, providing exceptional value at just $0.11 per clip:

• Critical cameraFixed rule — Set cameraFixed: true to prevent hallucinations (a hard requirement)
• Optimized for portrait — Native 9:16 support for vertical video formats
• Professional motion control — Fine-tune duration (5-10 seconds), orientation, and camera movement
• Atlas Cloud integration — Proven better value than alternatives like vast.ai

The Seedance implementation includes robust polling mechanisms to handle the 30-second to 2-minute generation time, making it reliable for production workflows.

Installation and Integration

Getting started is simple with Claude Code's plugin system:

/plugin marketplace add https://github.com/clickaround/marketplace
/plugin install image-generation@clickaround

For custom integrations, the TypeScript implementations are designed for easy portability:

• NanoBanana: Copy src/image-generation/ directory, install @google/generative-ai, set GOOGLE_GEMINI_API_KEY
• Seedance: Copy src/YTB-seedance-server/ files, install axios, set ATLAS_CLOUD_API_KEY

Real-World Use Cases

This skill pack powers several production applications:

• Poetry Shorts — Transforming poetic text into visually rich short videos
• Educational Content Platforms — Generating illustrative images for learning materials
• Book Promotion Tools — Creating cover art and chapter illustrations
• News Visualization — Converting news headlines into engaging visual representations

The combination of NanoBanana's image creation and Seedance's video conversion creates a complete pipeline from text to engaging multimedia content.