Skip to main content

Documentation Index

Fetch the complete documentation index at: https://developer.box.com/llms.txt

Use this file to discover all available pages before exploring further.

These tutorials walk you through each Box AI API endpoint with working code examples. For input limits, model behavior, and the ai_agent override system, see the . Start with the prerequisites to set up authentication, then follow the guide for the capability you need.

Prerequisites

Create a platform app, enable the AI scope, and generate a developer token.

Ask questions

Use POST /ai/ask to query files in single or multi-item mode. Includes dialogue history and citation examples.

Generate text

Use POST /ai/text_gen to create content from file-based prompts with iterative refinement through dialogue history.

Override model configuration

Change the AI model, prompt template, or LLM parameters to customize results for your use case.

Extract use cases

Explore industry-specific extraction scenarios for sales, legal, HR, financial services, and more.

Extract metadata (freeform)

Use POST /ai/extract with natural language prompts to pull key-value pairs from documents.

Extract metadata (structured)

Use POST /ai/extract_structured with metadata templates or field definitions. Includes OCR and Enhanced Extract Agent examples.