DocuProx Core Features - Templates & Integrations Guide

A DocuProx Template is a user-defined blueprint that teaches the DocuProx AI how to recognize, locate, and extract specific pieces of information from a particular type of document.

Key aspects of a DocuProx Template:

Folders

Organize and categorize your templates for better management. Create folders to group templates by document type, department, or any custom classification system that suits your workflow.

Reference Image Mode

Choose between two extraction approaches based on your document characteristics:

With Reference Image

Use this mode when you have a standard document format and want to guide the AI with a reference image. In this mode, you can define:

VALUE The actual value present in the reference document that will be used to identify label-value pair based extraction from actual documents at runtime

RECTANGLE A bounded region (numbered for identification) that defines the area where the target value is located, helping the system know exactly which region to extract data from

PROMPT Natural language instructions that tell the system how to identify and extract the target value

Without Reference Image

Use this mode for documents that don't follow a standard format and vary for each type. This approach relies entirely on prompt-based extraction using natural language descriptions to guide the AI.

Custom Instructions (Optional)

For specialized documents that the AI hasn't been extensively trained on, use this field to describe each aspect of your document in natural language. This helps the AI better understand the document structure and content.

JSON Structure Builder

A dynamic JSON builder that allows you to define the exact structure of your desired response format. You can:

• Add elements and nodes (objects) to create complex JSON structures
• Define target keys where extracted values will be populated at runtime when your system calls the backend API
• Configure VALUE, RECTANGLE, and PROMPT-based extraction methods within the JSON structure for precise data mapping

Test

After configuring your template, use the test feature to validate that your template returns the desired data in the correct format. You'll need to generate an API key to use this testing functionality, then upload sample documents to verify extraction accuracy before deploying to production.

DocuProx integrates seamlessly with your existing business systems and workflows through our robust API-first architecture.

Common Integration Scenarios

CRM Systems

Automatically process customer documents and populate CRM records (Salesforce, HubSpot).

Accounting Software

Extract invoice data and feed directly into QuickBooks, SAP, or ERP systems.

Document Management

Integrate with SharePoint, Google Drive, or custom document repositories.

Integration Methods

RESTful API

Direct API calls for real-time document processing and data extraction.

Webhooks

Real-time notifications and event-driven processing for automated workflows.

SDKs & Libraries

Pre-built libraries for popular programming languages (Python, JavaScript, Java).

Example Integration Workflow

Document Upload

User uploads document to your system

API Call

Your system calls DocuProx API with document

Data Extraction

DocuProx processes and extracts data

Data Storage

Structured data stored in your database