Image Generation Through Claude Code
I tell Claude Code what image I want and it makes it through Nano Banana. It prompts, generates, and checks. Public on GitHub.
This app is open on GitHub. Fork it, use it and build on it. I like giving it to the community.
About the project
When I'm building a site or making creatives, I need images. Instead of clicking around in a tool, I tell Claude Code what I want and it handles it through Nano Banana (Gemini 3 Pro Image).
It writes the prompt itself, generates the image, and checks whether the result is worth keeping.
Under the hood it's a connector to the Nano Banana API and a Claude Code skill. It does batches from JSON job files, reference images for multimodal input, retries with exponential backoff, and output in PNG and WebP.
It's public on GitHub, so go ahead and fork it, use it, and build on it.
What you get
- I say what I want and Claude Code makes the image.
- Batch: dozens of images from a single JSON job.
- It checks quality itself and retries bad attempts.
- Public on GitHub, fork it for yourself.
How it works (under the hood)
- A connector to the Nano Banana (Gemini 3 Pro Image) API.
- A Claude Code skill, batches from JSON job files.
- Reference images, retries with exponential backoff.
- Output in PNG and WebP.
Built with
Want to build like this too?
I teach building apps without programming in the AI First course.
More from Tool
Shoutbase: Testimonials Manager
An internal tool for managing testimonials: a queue of who to ask, a collection of what we've gathered, and an overview of where we use them.
Toucan: A Shared Second Brain for the Team
A shared cloud knowledge base that the team and its AI agents draw from, a "second brain" connected to my local context1.
Running a Project's Marketing Through Claude Code
A marketing team as a set of Claude Code skills: a director, department managers, and specialists that run a project alongside me.