Get your current user information and check connection status
Start the Box application authorization process
Search for files in Box with options for query, file extensions, search locations, and folder scope
Read the text content of a Box file
Ask Box AI about a file
Query Box AI using multiple files
Ask Box AI about a hub
Locate a folder by name
Extract data from a file using AI
List folder contents
Create, update, or delete folders in Box
Upload a file to Box from a local filesystem path
Upload content as a file to Box
Download a file from Box
The Box API Integration provides a powerful interface for working with Box content and leveraging Box's AI capabilities. This MCP server enables seamless interaction with Box files and folders, allowing for searching, reading, creating, updating, and deleting content. It also provides access to Box's AI features for extracting data, asking questions about files, and working with Box Hubs. With this integration, you can programmatically manage your Box content, extract text from documents, leverage AI to analyze file contents, and organize your data efficiently. The server handles authentication with Box and provides a comprehensive set of tools that make it easy to incorporate Box functionality into your AI workflows.
The Box API Integration MCP server provides a comprehensive set of tools for interacting with Box content and AI capabilities. This server allows you to seamlessly integrate Box functionality into your AI workflows, enabling you to search for files, read content, extract data using AI, and manage folders and files.
To install and use the Box API Integration MCP server, follow these steps:
Clone the repository:
git clone https://github.com/box-community/mcp-server-box.git
cd mcp-server-box
Set up a Python environment (Python 3.9+ recommended):
python -m venv venv
source venv/bin/activate # On Windows: venv\Scripts\activate
Install dependencies:
pip install -e .
Configure your Box application:
Set up environment variables for authentication:
BOX_CLIENT_ID=your_client_id
BOX_CLIENT_SECRET=your_client_secret
Start the MCP server:
python -m src.main
Add the server configuration to your AI client (Claude, Cursor, etc.):
"mcpServers": {
"box-api": {
"command": "python",
"args": [
"-m",
"src.main"
],
"cwd": "/path/to/mcp-server-box"
}
}
When you first use the Box API Integration, you'll need to authenticate with Box:
box_authorize_app_tool to start the authorization processbox_who_am_i to verify your connection statusTo search for files in Box:
Use the box_search_tool to find PDF files containing "quarterly report"
To read the content of a file:
Use box_read_tool with the file ID to extract the text content
To ask questions about a file using Box AI:
Use box_ask_ai_tool with the file ID and a question like "Summarize the main points of this document"
To create a new folder:
Use box_manage_folder_tool with action "create", name "Project Documents", and parent_id "0" (for root folder)
To upload a file to Box:
Use box_upload_file_from_path_tool with the local file path and destination folder ID
box_authorize_app_tool