Nano Banana MCP

mcp-server-nano-banana is an MCP server that connects Claude Desktop and other Model Context Protocol clients to the Nanana AI image generation service powered by Google Gemini's Nano-Banana model. Built for developers, designers, and creative workflows, this server enables AI assistants to produce visuals directly inside chat interfaces. Users can execute generative tasks via two primary capabilities: generating brand-new images from descriptive text prompts, and transforming existing imagery through image-to-image styling instructions. The integration manages authentication using personal Nanana API tokens and exposes dedicated tools for image synthesis. Instead of jumping between external web applications and development environments, users can command their MCP client to draft mockups, generate character concepts, or apply artistic styles like oil painting to existing assets. By establishing a direct bridge to Nanana AI APIs, it centralizes creative image manipulation and generative design into standard conversational agent pipelines.

Category: Design, Media & Creative

Tags: gemini, Generative AI, image generation, multimodal

Visit Nano Banana MCP

How to install and configure Nano Banana MCP

Follow these steps to install and configure mcp-server-nano-banana for Claude Desktop: 1. Install the package globally via npm: bash npm install -g @nanana-ai/mcp-server-nano-banana 2. Obtain an API token from your Nanana AI account dashboard under API Access. 3. Open your Claude Desktop configuration file (claude_desktop_config.json) and add the server configuration: json { "mcpServers": { "nanana": { "command": "npx", "args": ["-y", "@nanana-ai/mcp-server-nano-banana"], "env": { "NANANA_API_TOKEN": "your-api-token-here" } } } } 4. Save the configuration file and restart Claude Desktop to load the new tools.

What you can do with Nano Banana MCP

• Creating custom illustrations and marketing artwork directly from textual prompt descriptions inside your conversational assistant interface. - Applying artistic modifications and style transfers to between one and nine existing images supplied by publicly accessible URLs. - Generating placeholder graphics, character concepts, and visual assets rapidly during interactive software development and interface design sessions. - Automating iterative image modifications by requesting targeted prompt refinements through an MCP-compatible agent or Claude Desktop.

Key facts

• https://github.com/nanana-app/mcp-server-nano-banana
• Design, Media & Creative
• gemini, Generative AI, image generation, multimodal

Part of MCP Servers

How do I install mcp-server-nano-banana?

You can install it globally using the npm command npm install -g @nanana-ai/mcp-server-nano-banana. After installation, register the server inside your Claude Desktop configuration file using npx to launch @nanana-ai/mcp-server-nano-banana, and make sure to pass your NANANA_API_TOKEN environment variable.

What tools are available in mcp-server-nano-banana?

The server provides two primary tools. The text_to_image tool generates images from descriptive text prompts. The image_to_image tool accepts an array of one to nine image URLs alongside a transformation prompt to modify existing visuals into new styles or variations.

Which MCP clients work with mcp-server-nano-banana?

It is designed to work with Claude Desktop and any other MCP-compatible client that supports standard Model Context Protocol server definitions running over stdio execution environments.

Is mcp-server-nano-banana open source?

Yes, mcp-server-nano-banana is open source and released under the MIT license on GitHub.

How does pricing work for mcp-server-nano-banana?

Generating images with the Nano-Banana model consumes credits from your Nanana AI account. You can monitor your current credit balance and purchase additional credits directly in your Nanana account dashboard.