Image Generation Skills for Claude Code — Gemini NanoBanana & Seedance

AI-powered image generation is transforming how creators build visual content. With ClickAround Creator Pro's image-generation skill pack, you get two powerful capabilities in one open-source package: Gemini NanoBanana for AI image creation and Seedance for image-to-video conversion. This article walks through what these skills do, how they work, and why they're valuable for developers and content creators.

Gemini NanoBanana: AI Image Generation

The image-nanobanana skill enables AI image generation using Google's Gemini models. It's designed specifically for creating vertical (9:16) images optimized for Shorts, Reels, and other social media formats. The skill automatically applies style prefixes like "Ghibli" or "Educational" to your prompts, ensuring consistent artistic direction across your generated images.

Key features of the NanoBanana image generation:

• Vertical format optimization: Generates images in 9:16 aspect ratio by default, perfect for mobile-first content
• Style prefixing: Automatically adds artistic style descriptors to your prompts for consistent results
• Character reference mode: Supports character consistency across multiple image generations
• Cost-efficient: Minimizes API calls with intelligent batching and configuration options

This skill integrates seamlessly with TypeScript projects and requires only the @google/generative-ai SDK and your Google Gemini API key. The implementation is clean and well-documented, making it easy to integrate into existing projects or adapt for new use cases.

Seedance: Image-to-Video Conversion

While static images are powerful, motion adds another dimension to storytelling. The image-seedance skill bridges this gap by converting your AI-generated images into short video clips using Seedance 1.5 technology via Atlas Cloud API.

Key features of the Seedance image-to-video conversion:

• Cost-effective: ~$0.11 per clip, making it significantly more affordable than alternatives
• Optimized for vertical content: Designed specifically for 9:16 portrait orientation
• Stable motion: Requires cameraFixed: true to prevent hallucinations and ensure stable results
• Fast processing: Generates clips in 30 seconds to 2 minutes depending on complexity

The Seedance integration uses a simple TypeScript interface that's easy to implement in any project. You just need the axios library and your Atlas Cloud API key. The implementation includes proper error handling and polling mechanisms to manage the asynchronous nature of video generation.

How These Skills Work Together

The true power of the image-generation pack emerges when you combine both skills in a workflow. Here's how it works:

1. Generate base images: Use NanoBanana to create a series of thematically consistent images
2. Add motion: Convert each image to a short video clip using Seedance
3. Assemble final content: Combine clips into longer videos or use individually as social media posts

This end-to-end workflow eliminates the need for separate tools and APIs, providing a unified solution that's both cost-effective and developer-friendly. The TypeScript implementation ensures type safety and makes debugging straightforward.

Getting Started

To install the image-generation skill pack in Claude Code, simply run:

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

You'll need to set up two environment variables:

• GOOGLE_GEMINI_API_KEY for NanoBanana image generation
• ATLAS_CLOUD_API_KEY for Seedance video generation

Both skills are completely open source under the MIT license, meaning you can use them freely in personal and commercial projects without restrictions.