Skip to main content
Glama
Sign In
Nano Banana MCP Server

Nano Banana MCP Server

by NH-5
OverviewSchemaSourceRelated ServersDiscussionsAdmin
TypeScript
Hybrid

Nano Banana MCP Server

MCP服务器，用于通过Google AI Studio API调用Gemini Nano Banana Pro (gemini-3-pro-image-preview) 图像生成模型。

版本: 1.0.0

功能

  • 调用Google Gemini Nano Banana模型生成图像

  • 支持通过MCP协议在opencode等客户端中使用

  • 生成图片会保存到本地文件

Related MCP server: gemini-image-mcp

安装

npm install
npm run build

配置

  1. 复制 .env.example 为 .env：

cp .env.example .env

  1. 在 .env 中设置你的Google API Key：

GOOGLE_API_KEY=your_api_key_here

环境变量

变量

说明

默认值

GOOGLE_API_KEY

Google AI Studio API密钥

(必填)

HTTPS_PROXY

HTTPS代理地址

系统代理

运行

npm start

在opencode中使用

在opencode的mcp配置中添加：

{
  "mcp": {
    "nanobanana": {
      "type": "local",
      "command": ["node", "/path/to/nanobanana-mcp-server/dist/index.js"],
      "env": {
        "GOOGLE_API_KEY": "your_api_key"
      },
      "enabled": true
    }
  }
}

工具

generate_image

生成图像

参数：

  • prompt (required): 图像描述文本

  • output_dir: 图片保存目录 (默认当前目录)

示例：

用nano banana画一只可爱的猫咪

Available Tools

1 tool
generate_imageC

Generate images using Google Gemini Nano Banana Pro (gemini-3-pro-image-preview) model

Inputs: prompt (required), output_dir.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    Deploy Server
    One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

    TDQS

    B3.1/5.0

    Scored across 1 tool

    Disambiguation5/5

    With only one tool, there is no possibility of ambiguity or overlap between tools. The single tool 'generate_image' has a clearly distinct purpose that cannot be confused with any other tool in this set.

    Naming Consistency5/5

    The single tool name 'generate_image' follows a clear verb_noun pattern. With only one tool, there is perfect consistency as there are no other tools to compare against or create naming conflicts with.

    Tool Count2/5

    A single tool for an image generation server feels thin and under-scoped. While it might cover the core functionality, typical MCP servers for such domains would include additional tools like list_models, get_image_details, or variations of generation parameters. The count of 1 suggests limited capability for agents to perform related operations.

    Completeness2/5

    The tool surface is severely incomplete for an image generation domain. There is only a generation tool with no supporting operations like model listing, configuration management, image retrieval, or editing capabilities. This creates significant gaps that will likely cause agent failures when trying to perform comprehensive image-related tasks.

    Maintenance

    ActivityInactive
    ResponsivenessOutdated data

    Resources

    Related MCP Connectors

    Related MCP Servers