gemini-cli-mcp-server

gemini-cli-mcp-server

A MCP server wrapper for Google's Gemini CLI that enables AI assistants to perform searches, have conversations, list sessions, and analyze files using Gemini's capabilities.

Category
访问服务器

README

Gemini CLI MCP Server

CI

🇯🇵 日本語

A simple MCP server wrapper for Google's Gemini CLI that enables AI assistants to use Gemini's capabilities through the Model Context Protocol.

✨ Features

  • 4 Tools: googleSearch, chat, listSessions, analyzeFile
  • Session Persistence: Resume previous conversations with session IDs (supports custom IDs)
  • Internationalization: English and Japanese support
  • Cross-Platform: Windows, macOS, and Linux compatible

🚀 Quick Setup

With Claude Code

claude mcp add -s project gemini-cli -- npx gemini-cli-mcp-server --allow-npx

Manual Configuration

{
  "mcpServers": {
    "gemini-cli-mcp-server": {
      "command": "npx",
      "args": ["gemini-cli-mcp-server", "--allow-npx"]
    }
  }
}

🛠️ Available Tools

googleSearch

Performs a Google search using Gemini CLI.

Parameter Required Description
query The search query
limit Maximum number of results
raw Return structured results with URLs
model Gemini model (default: gemini-2.5-pro)

chat

Have a conversation with Gemini.

Parameter Required Description
prompt The conversation prompt
sessionId Resume a previous session (supports custom IDs like "my-task")
model Gemini model (default: gemini-2.5-pro)

listSessions

Lists available Gemini CLI sessions. Returns session IDs that can be used with chat's sessionId parameter.

analyzeFile

Analyze files using Gemini's multimodal capabilities.

Parameter Required Description
filePath Absolute path to the file
prompt Additional analysis instructions
sessionId Resume a previous session (maintains context)
model Gemini model (default: gemini-2.5-pro)

Supported file types:

  • Images: PNG, JPG, JPEG, GIF, WEBP, SVG, BMP
  • Text: TXT, MD
  • Documents: PDF

💾 Session Management

You can use Custom Session IDs (Client IDs) to manage conversations. The server automatically maps these to Gemini CLI's internal session IDs.

{
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "name": "chat",
    "arguments": {
      "prompt": "Hello!",
      "sessionId": "my-task-1"
    }
  },
  "id": 1
}
  • Persistence: Mappings are stored in memory. They will be reset if the MCP server restarts (Gemini's internal history remains, but the link to your custom ID is lost).
  • Listing: Use listSessions to see active mappings.

🌐 Language Settings

Tool descriptions and error messages support multiple languages:

# Japanese
MCP_LANGUAGE=ja npx gemini-cli-mcp-server --allow-npx

# English (default)
MCP_LANGUAGE=en npx gemini-cli-mcp-server --allow-npx

System locale (e.g., LANG=ja_JP.UTF-8) is also auto-detected.

📝 Development

# Clone and install
git clone https://github.com/Cinnamobot/gemini-cli-mcp-server
cd gemini-cli-mcp-server
bun install

# Run tests
bun test

# Build
bun run build

🙏 Credits

This project is based on choplin/mcp-gemini-cli.

Additions from fork

  • listSessions tool: List available Gemini CLI sessions
  • Session persistence: Resume conversations with sessionId parameter
  • Internationalization: Japanese/English support
  • Windows compatibility: Custom findExecutable function (no which/where dependency)
  • CI/CD: GitHub Actions for automated testing and building

📄 License

MIT License - see LICENSE file for details.

🔗 Related Links

推荐服务器

Baidu Map

Baidu Map

百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。

官方
精选
JavaScript
Playwright MCP Server

Playwright MCP Server

一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。

官方
精选
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。

官方
精选
本地
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。

官方
精选
本地
TypeScript
VeyraX

VeyraX

一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。

官方
精选
本地
graphlit-mcp-server

graphlit-mcp-server

模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。

官方
精选
TypeScript
Kagi MCP Server

Kagi MCP Server

一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。

官方
精选
Python
e2b-mcp-server

e2b-mcp-server

使用 MCP 通过 e2b 运行代码。

官方
精选
Neon MCP Server

Neon MCP Server

用于与 Neon 管理 API 和数据库交互的 MCP 服务器

官方
精选
Exa MCP Server

Exa MCP Server

模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。

官方
精选