Name: Nano Banana MCP
Availability: InStock
Author: pierceboggan

What it does

Nano Banana MCP lets you generate images from text prompts using Google Gemini 2.5 Flash, and renders the results as an interactive image viewer directly inside your MCP Apps-capable client. You can also pass reference images to restyle, edit, or combine them — all without leaving the conversation.

Key features

Text-to-image generation via Google Gemini 2.5 Flash — describe what you want and get a visual result inline
Image editing with references — provide existing images and a prompt to restyle, modify, or merge them
MCP Apps inline rendering — generated images appear as an interactive viewer inside VS Code Copilot, Claude Desktop, Goose, and other MCP Apps-capable clients
Save to disk — optionally persist generated images to a local directory with custom filenames
Cross-platform — works on any system with Node.js and a Google AI Studio API key

Installation

Claude Desktop

Add to claude_desktop_config.json:

{
  "mcpServers": {
    "nano-banana": {
      "command": "npx",
      "args": ["-y", "nano-banana-mcp"],
      "env": {
        "GOOGLE_API_KEY": "your-google-api-key"
      }
    }
  }
}

VS Code Copilot

Add to your mcp.json (User or workspace):

{
  "servers": {
    "nano-banana": {
      "type": "stdio",
      "command": "npx",
      "args": ["-y", "nano-banana-mcp"],
      "env": {
        "GOOGLE_API_KEY": "${input:google-api-key}"
      }
    }
  },
  "inputs": [
    {
      "id": "google-api-key",
      "type": "promptString",
      "description": "Google AI Studio key",
      "password": true
    }
  ]
}