Nano Banana

MCP server for Nanana AI image generation powered by Google Gemini's nano banana

javascript

6

GitHub Stars

javascript

Language

4 months ago

First Indexed

3 weeks ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "nanana-app-mcp-server-nano-banana": {
      "command": "npx",
      "args": [
        "-y",
        "@nanana-ai/mcp-server-nano-banana"
      ],
      "env": {
        "NANANA_API_URL": "http://localhost:3000",
        "NANANA_API_TOKEN": "your-api-token-here"
      }
    }
  }
}

You run an MCP server to enable clients like Claude Desktop to generate and transform images using Nanana AI’s nano banana model. This server bridges requests from MCP-compatible clients to Nanana’s image-generation capabilities, delivering practical, scriptable image creation and editing.

How to use

Start by installing the MCP server globally, then configure your client to connect through a local or remote MCP endpoint. Use the standardized tools to generate new images from prompts or transform existing images with descriptive prompts.

How to install

Prerequisites: you need Node.js and npm installed on your system.

Install the MCP server package globally.

Configuration and notes

API tokens are required to access Nanana AI services. Configure your MCP to pass your token securely to the server. If you plan to run a local instance for testing, you can point the client at a local URL set through environment variables.

Environment variables you will use include NANANA_API_TOKEN for authentication and NANANA_API_URL to override the default API endpoint.

Examples and troubleshooting

If you encounter authentication errors, double-check that your NANANA_API_TOKEN is valid and that NANANA_API_URL is reachable. Ensure your MCP client is configured to point at the correct MCP server entry point.

Available tools

text_to_image

Generates a new image from a text prompt. Provide a descriptive prompt to guide the image creation.

image_to_image

Transforms existing images by applying a new prompt to modify style, composition, or attributes. Provide image URLs and a transformation prompt.