Agent skill for AI image generation and editing using Nano-Banana MCP server and Google Gemini

Nano-Banana Skill 🍌🧠

Agent skill for AI image generation and editing — teaches Claude Code, Codex, and other AI agents how to use the mcp-nanobanana MCP server like a pro.

This skill is the companion to the Nano-Banana MCP server. It provides smart routing, prompt enhancement, bilingual support (PT-BR & EN), and natural language mapping for aspect ratios and resolutions.

✨ What It Does

• 🎨 Smart Tool Routing — Automatically picks the right tool based on context (generate, edit, or refine)
• ✍️ Prompt Enhancement — Transforms simple descriptions into detailed, high-quality prompts
• 🌎 Bilingual Support — Understands commands in Portuguese and English
• 📐 Natural Language Mapping — "widescreen em 4K" or "square in high res" → correct parameters
• 🔄 Multi-Turn Editing — Understands when to continue the current image session or start fresh
• 🧩 Reference Images — Knows how to route visual references into generation and editing tools
• 💬 Conversation Patterns — Knows when to ask for aspect ratio/resolution and when to skip

📋 Prerequisites

Before installing the skill, you need:

1. Claude API key (free) — get yours at Google AI Studio

2. mcp-nanobanana MCP server installed and configured in your client.

   For complete installation instructions (Claude Code, Codex, Gemini, Windsurf, and others), see the mcp-nanobanana README.

   Quick setup for Claude Code:

   Claude MCP add nanobanana -e GEMINI_API_KEY="your-gemini-api-key-here" -- npx mcp-nanobanana
   

   Quick setup for Codex:

   codex mcp add nanobanana --env GEMINI_API_KEY="your-gemini-api-key-here" -- npx mcp-nanobanana
   

   Other clients — add this to your MCP settings file:

   {
     "mcpServers": {
       "nanobanana": {
         "command