Box MCP Server
Enables AI agents to manage Box files, folders, collaborations, and more via the Box API using tools for search, upload, and metadata operations.
README
🚀 Model Context Protocol API (MCP Server)
Welcome to the official source code repository for the Model Context Protocol Tutorial This is designed to help you get started with Model Context Protocol (MCP) — a powerful future-ready technology to connect external context to your AI application. MCP is an open protocol that standardizes how applications provide context to large language models (LLMs). Think of MCP like a USB-C port for AI applications. Just as USB-C provides a standardized way to connect your devices to various peripherals and accessories, MCP provides a standardized way to connect AI models to different data sources and tools. MCP enables you to build agents and complex workflows on top of LLMs and connects your models with the world. 🎬 Watch the full tutorial:
👉 Model Context Protocol Tutorial on YouTube
📚 What You’ll Learn
- ✅ What is MCP and how it works
- ✅ What is MCP Server, why we use it and how to use it
- ✅ Real life code example
🎯 Prerequisites for Learning MCP
To get the most out of this curriculum, you should have:
- Basic knowledge of programming in at least one of the following languages: C#, Java, JavaScript, Python, or TypeScript
- Understanding of client-server model and APIs
- Familiarity with REST and HTTP concepts
- (Optional) Background in AI/ML concepts
📚 Study Guide & Resources
This repository includes several resources to help you navigate and learn effectively: https://github.com/microsoft/mcp-for-beginners.git
Official Website
https://modelcontextprotocol.io/docs/getting-started/intro
Box MCP Server
Quick Start
Clone the repository:
git clone https://github.com/box-community/mcp-server-box.git
cd mcp-server-box
Optional but recommended uv installation for virtual environment and dependency management:
Homebrew (macOS)
brew install uv
WinGet (Windows)
winget install --id=astral-sh.uv -e
On macOS and Linux
curl -LsSf https://astral.sh/uv/install.sh | sh
On Windows
powershell -ExecutionPolicy ByPass -c "irm https://astral.sh/uv/install.ps1 | iex"
Set up the virtual environment and install dependencies:
uv sync
Set environment variables:
Set the following environment variables for Box authentication in a .env file or your system environment:
Using OAuth2.0 with a Box App
BOX_CLIENT_ID = YOUR_CLIENT_ID
BOX_CLIENT_SECRET = YOUR_CLIENT_SECRET
BOX_REDIRECT_URL = http://localhost:8000/callback
BOX_MCP_SERVER_AUTH_TOKEN = YOUR_BOX_MCP_SERVER_AUTH_TOKEN
Using CCG with a Box App
BOX_CLIENT_ID = YOUR_CLIENT_ID
BOX_CLIENT_SECRET = YOUR_CLIENT_SECRET
BOX_SUBJECT_TYPE = user_or_enterprise
BOX_SUBJECT_ID = YOUR_USER_OR_ENTERPRISE_ID
BOX_MCP_SERVER_AUTH_TOKEN = YOUR_BOX_MCP_SERVER_AUTH_TOKEN
Note: The
BOX_MCP_SERVER_AUTH_TOKENis the token used to authenticate requests to the Box MCP server. You can generate this token.
Run the MCP server in STDIO mode:
uv run src/mcp_server_box.py
Box Community MCP Server Tools
Below is a summary of the available tools:
| Tools available | Description |
|---|---|
| box_tools_ai | AI-powered file and hub queries |
| box_tools_collaboration | Manage file/folder collaborations |
| box_tools_docgen | Document generation and template management |
| box_tools_files | File operations (read, upload, download) |
| box_tools_folders | Folder operations (list, create, delete, update) |
| box_tools_generic | Generic Box API utilities |
| box_tools_groups | Group management and queries |
| box_tools_metadata | Metadata template and instance management |
| box_tools_search | Search files and folders |
| box_tools_shared_links | Shared link management for files/folders/web-links |
| box_tools_users | User management and queries |
| box_tools_web_link | Web link creation and management |
Box Community MCP Server Operations Details
Command line interface parameters
To run the MCP server with specific configurations, you can use the following command line parameters:
uv run src/mcp_server_box.py --help
usage: mcp_server_box.py [-h] [--transport {stdio,sse,streamable-http}] [--host HOST]
[--port PORT] [--box-auth {oauth,ccg}] [--no-mcp-server-auth]
Box Community MCP Server
options:
-h, --help show this help message and exit
--transport {stdio,sse,streamable-http}
Transport type (default: stdio)
--host HOST Host for SSE/HTTP transport (default: 0.0.0.0)
--port PORT Port for SSE/HTTP transport (default: 8000)
--box-auth {oauth,ccg}
Authentication type for Box API (default: oauth)
--no-mcp-server-auth Disable authentication (for development only)
Claude Desktop Configuration
Edit your claude_desktop_config.json:
code ~/Library/Application\ Support/Claude/claude_desktop_config.json
Add the configuration:
{
"mcpServers": {
"mcp-server-box": {
"command": "uv",
"args": [
"--directory",
"/path/to/mcp-server-box",
"run",
"src/mcp_server_box.py"
]
}
}
}
Restart Claude if it is running.
Cursor Configuration
Cursor supports MCP servers through its configuration file. Here's how to set it up:
The Cursor MCP configuration file is located at:
- macOS/Linux:
~/.cursor/config.jsonor~/.config/cursor/config.json - Windows:
%APPDATA%\Cursor\config.json
Add the MCP Server Configuration: STDIO Transport
Edit your Cursor configuration file and add the following under the mcpServers section:
{
"mcpServers": {
"mcp-server-box": {
"command": "uv",
"args": [
"--directory",
"/path/to/mcp-server-box",
"run",
"src/mcp_server_box.py"
],
"env": {
"BOX_CLIENT_ID": "YOUR_CLIENT_ID",
"BOX_CLIENT_SECRET": "YOUR_CLIENT_SECRET",
"BOX_REDIRECT_URL": "http://localhost:8000/callback"
}
}
}
}
推荐服务器
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。
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。
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。
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。
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。
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。
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。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
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。