tasopen/mcp-alphabanana

Overview

A local MCP server that generates image assets using Google Gemini AI models (Nano Banana 2 / Pro). It supports transparency and resizing of generated images. The server runs locally and exposes tools for integration into development workflows.

Best for

Best for
Developers who need local image generation with transparency and resize support using Gemini models

Use cases

• Generate transparent PNG icons for UI components
• Create resized image assets for web or mobile projects
• Prototype visuals using Gemini models without external API calls

How to use

Install

npx -y @tasopen/mcp-alphabanana

Tools exposed

• output_resolution
• output_format
• grounding_type
• thinking_mode
• include_thoughts
• include_metadata

Tested with

Claude Desktop, Cursor, VS Code

Notes

A local MCP server that generates image assets using Google Gemini AI models (Nano Banana 2 / Pro). It supports transparency and resizing of generated images. The server runs locally and exposes tools for integration into development workflows.

6 stars on GitHub. Last updated 2026-04-06. Licensed MIT.

Use cases

• Generate transparent PNG icons for UI components
• Create resized image assets for web or mobile projects
• Prototype visuals using Gemini models without external API calls

Pros

• Runs locally for privacy and offline capability
• Supports transparency and resizing out of the box
• Leverages Gemini models for image generation

Cons

• Low community adoption (6 stars) indicates early stage
• Requires a Google Gemini API key to function
• Limited to specific Gemini models (Nano Banana 2 / Pro)

Indexed from awesome-mcp-servers-punkpeye and enriched against its public facts.