Box AI Extract Quickstart

A Python application that demonstrates how to use Box AI to extract structured metadata from files and attach it back to the file in Box. This quickstart guide will walk you through these steps.

Prerequisites

• Python 3.11+
• Box Developer Account
• Box App with appropriate permissions

Setup

1. Clone the repository

   git clone https://github.com/box-community/box-ai-extract-quickstart.git
   cd box-ai-extract-quickstart

2. Create virtual environment

   python -m venv venv
   source venv/bin/activate  # On Windows: venv\Scripts\activate

3. Install dependencies

   pip install python-dotenv "boxsdk>=10"

4. Configure environment variables

   cp .env.template .env

   Edit .env and add your values:

   • BOX_DEVELOPER_TOKEN: Your Box developer token
   • BOX_FILE_ID: ID of the file to extract metadata from
   • BOX_METADATA_TEMPLATE_KEY: Key of your metadata template

Usage

Run the extraction script:

python extract.py

The script will:

1. Authenticate with Box using your developer token
2. Extract structured metadata from the specified file using Box AI
3. Attach the extracted metadata to the file

Box Setup Requirements

1. Create a Box App with Client Credentials Grant authentication
2. Create a metadata template in your Box enterprise
3. Ensure your app has permissions to read files and manage metadata
4. Upload a file to Box and note its file ID

Project Structure

• extract.py - Main application script
• .env.template - Environment variables template
• .env - Your environment variables (not tracked in git)
• .gitignore - Git ignore rules for Python projects