Getting Started with DocuProx
Get up and running with DocuProx AI document extraction in minutes. This guide covers everything from account setup to your first successful API call.
Introduction
DocuProx is a powerful, AI-driven SaaS platform designed to intelligently extract structured data from images and PDF documents like invoices, passports, and government IDs. Built with an API-first approach, DocuProx provides a robust RESTful API specifically designed for seamless integration into your existing systems. Using configurable templates (value-based, annotation-based, and prompt-based extraction methods), DocuProx eliminates manual data entry, enhances accuracy, and accelerates your business processes by transforming unstructured document data into ready-to-use JSON format that can be directly consumed by your applications.
Getting Started with Templates
Getting started with Templates is straightforward and requires just a few simple steps:
Sign In
Create your DocuProx account and access the platform
Don't have access? Request access through our signup request form
Generate API Key
Create a secure API key for authentication
Configure Templates
Set up templates using value-based, annotation-based, or prompt-based extraction methods
Define JSON Structure
Specify the exact JSON format for your extracted data
Test Template
Upload sample images to validate template output, check accuracy, and adjust template configuration if needed
Integrate Your System
Use your template ID along with images or PDFs to extract structured data via API calls
Getting Started with Agentic-AI
Getting started with Agentic-AI is straightforward and requires just a few simple steps:
Sign In
Create your DocuProx account and access the platform
Don't have access? Request access through our signup request form
Generate API Key
Navigate to the API Keys section in your dashboard and create a secure API key for authenticating requests to the /v1/process-agent endpoint
Define Prompt & Call API
Build your request payload with document_type, custom_instructions, and prompt_json fields to dynamically define what data to extract — no template setup needed
Send documents directly via the /v1/process-agent endpoint with your prompts and instructions
Integrate & Automate
Connect the Agentic-AI endpoint to your existing systems (Salesforce, ServiceNow, custom apps) for autonomous, prompt-driven document processing at scale
API Integration Flow
Customer System
Your existing business systems, such as Salesforce, ServiceNow, or custom applications.
DocuProx API
Handles secure and scalable document processing and data extraction via RESTful endpoints.
What this documentation covers:
- Platform Overview: Understanding what DocuProx is and its core capabilities.
- Quick Start Guide: Essential steps to get you up and running quickly, including API key generation, template configuration, and testing.
- Core Topics: Detailed explanations of how DocuProx works, template management, integration with your applications, and monitoring your usage.
- Resources: Information on SDKs, libraries, and examples to aid in your development.
- FAQs: Answers to frequently asked questions and troubleshooting tips.
Quick Start
This section will guide you through the essential steps to quickly get started with DocuProx.
With Templates
1 Logging In and Navigating the Dashboard
When you access DocuProx, first you will need to log in with your Email ID and Password. After successful login, you will see your Dashboard and the left sidebar navigation.
Sidebar Navigation
The left sidebar provides quick access to the main areas of the application:
- • Dashboard
- • Templates
- • Manage Users
- • Profile
- • Settings
Dashboard Overview
The Dashboard provides a quick summary of your account's performance and usage. Here you will find real-time metrics and analytics, helping you monitor your document processing activities.
Key Dashboard Metrics:
Usage Summary
- • API Requests count
- • Credits Used/Remaining
- • Success Rate percentage
Analytics Filters
- • Last 7 days
- • Last 30 days
- • Last 90 days
Graphs & Trends
- • Requests Over Time
- • Success & Error Rates
- • Response Time Trend
2 How to Generate an API Key
Your API Key is crucial for securely integrating DocuProx with your applications and for testing templates.
Security Warning
For security reasons, this key will only be fully visible once. If you lose it, you will need to generate a new one. You can also revoke existing keys from here.
3 How to Configure a Template
Templates are blueprints that define how DocuProx extracts data from your documents.
Template Creation Steps:
Data Extraction Methods:
Label-Value Extraction (Prompt-based)
Highly effective when information has associated labels (e.g., "Name:", "Date:", "Invoice Total:").
Annotation-based Extraction (Visual)
Powerful for extracting data from specific, visually defined document areas, even without clear textual labels.
4 How to Test a Template
Thorough template testing is crucial to ensure accurate data extraction prior to large-scale deployment.
5 How to Integrate DocuProx with Your Application
DocuProx is designed for seamless integration into virtually any application, system, or workflow. Its API-first design ensures maximum flexibility.
Use Your API Key
The API key generated in Step 2 is your authentication token for all API requests.
Refer to API Documentation
For detailed integration instructions, including API endpoints, request/response formats, and code examples.
Utilize SDKs/Libraries
DocuProx offers SDKs or libraries in popular programming languages to simplify integration.
With Agentic-AI (Without Templates)
The Agentic-AI approach lets you process documents dynamically using the /v1/process-agent endpoint — no template setup required. Define your extraction prompts on-the-fly via API.
Prerequisites: Complete Steps 1-2 from the Templates section above (Login and API Key Generation). The same API key works for both template-based and Agentic-AI requests.
1 Build Your Process-Agent Request
With Agentic-AI, you skip template configuration entirely. Instead, define your extraction logic directly in the API payload using prompts and instructions.
Payload Parameters:
| Field | Type | Description |
|---|---|---|
| document_type | string | Type of document (e.g., "invoice", "receipt", "passport", "contract") |
| custom_instructions | string (optional) | Natural language instructions to guide extraction for specialized documents |
| prompt_json | object | JSON object defining fields to extract with prompt-based instructions per field |
| static_values | string (optional) | JSON string of static key-value pairs to include in the response |
Example Payload:
{
"document_type": "passport",
"custom_instructions": "Extract all personal information from this travel document",
"prompt_json": {
"passport_number": "Extract the passport number from the document",
"full_name": "Extract the full name of the passport holder",
"date_of_birth": "Extract the date of birth in YYYY-MM-DD format",
"expiry_date": "Extract the passport expiry date in YYYY-MM-DD format",
"nationality": "Extract the nationality/country of citizenship"
},
"static_values": "{ \"processed_by\": \"DocuProx Agentic-AI\" }"
}
2 Send the API Request
Submit your document along with the extraction payload to the Process Agent endpoint.
document_type to describe your document category.
prompt_json with field names as keys and extraction prompts as values.
custom_instructions for specialized document handling.
POST request to /v1/process-agent with your API key in the x-auth or Authorization header.
Example cURL Request:
curl -X POST "https://api.docuprox.com/v1/process-agent" \
-H "x-auth: YOUR_API_KEY" \
-H "Content-Type: multipart/form-data" \
-F "file=@/path/to/document.pdf" \
-F "document_type=invoice" \
-F 'prompt_json={"invoice_number": "Extract the invoice number", "total_amount": "Extract the total amount due"}'
3 Process the Response
The API returns structured JSON data based on your prompt_json field definitions.
Example Response:
{
"success": true,
"data": {
"passport_number": "AB1234567",
"full_name": "John Michael Smith",
"date_of_birth": "1985-03-15",
"expiry_date": "2028-06-20",
"nationality": "United States",
"processed_by": "DocuProx Agentic-AI"
},
"confidence_scores": {
"passport_number": 0.98,
"full_name": 0.97,
"date_of_birth": 0.95,
"expiry_date": 0.96,
"nationality": 0.99
},
"processing_time_ms": 1250
}
4 Integrate & Automate
Connect the Agentic-AI endpoint to your existing systems for autonomous, prompt-driven document processing at scale.
Dynamic Extraction
Define different prompt_json payloads per document type — no need to create or manage templates.
Workflow Integration
Connect to Zapier, n8n, Make, or custom webhooks for automated document processing pipelines.
SDK Support
Use our Node.js or Python SDKs to simplify integration with the Process Agent API.
When to Use Agentic-AI vs Templates
Use Agentic-AI for ad-hoc extractions, prototyping, or when document formats vary frequently. Use Templates for production workloads with consistent document formats where you want to fine-tune extraction accuracy.
Ready to explore more?
Now that you understand the basics, dive deeper into DocuProx features and capabilities.