Summary

This project provides a Claude Code skill for image generation and editing via the Gemini CLI's nanobanana extension. It targets Claude Code users needing a streamlined way to create diverse visuals, from text-to-image prompts to diagrams and photo restoration, within their coding environment.

How It Works

The skill integrates nanobanana, a Gemini CLI extension, as a callable function within Claude Code. Users interact via natural language prompts or commands, which the Gemini CLI processes. It leverages Google's Gemini models via API calls for image generation, editing, restoration, and diagram creation. This approach centralizes advanced image manipulation capabilities within Claude Code, simplifying workflows.

Quick Start & Requirements

• Installation:
  1. Gemini CLI: npm install -g @google/gemini-cli
  2. API Key: export NANOBANANA_GEMINI_API_KEY="your-api-key"
  3. nanobanana extension: gemini extensions install https://github.com/gemini-cli-extensions/nanobanana
  4. Skill: Copy SKILL.md to ~/.claude/skills/nano-banana/ or clone repo.
• Prerequisites: Gemini CLI (Node.js/npm), Gemini API Key (Google AI Studio), Claude Code environment.
• Output: Images saved to ./nanobanana-output/.
• Links: Gemini CLI extensions: https://github.com/gemini-cli-extensions/nanobanana

Highlighted Details

• Features: Text-to-image, editing, restoration, icon generation, diagram creation, pattern generation, sequential images.
• Commands: /generate, /edit, /restore, /icon, /diagram, /pattern, /story, /nanobanana (natural language).
• Options: --count=N (1-8 variations), --preview (auto-open), --styles="style1,style2", --seed=N (reproducibility).
• Models: Default gemini-2.5-flash-image (~$0.04/image); optional gemini-3-pro-image-preview for higher quality.

Maintenance & Community

Contributions are welcomed via issues or pull requests on the repository. No specific community channels (e.g., Discord, Slack) or dedicated roadmap links are provided.

Licensing & Compatibility

The project is licensed under the MIT license. This permissive license generally allows commercial use and integration into closed-source projects without significant restrictions.

Limitations & Caveats

The skill depends on the Claude Code environment. It also relies on external services: Gemini CLI and Google's Gemini API, which may incur costs or have usage limitations. Installation involves multiple steps across different tools.