PDF.co API icon

PDF.co API

Generate PDF, extract data from PDF, split PDF, merge PDF, convert PDF. Fill PDF forms, add text and images to pdf and much more with pdf.co!

AI Invoice Parser

AI-generated

Overview

This node performs AI-powered invoice parsing from a PDF file accessible via a URL. It extracts structured data from invoices, which is useful for automating data entry, accounting, and financial workflows. For example, it can extract invoice numbers, dates, totals, and custom fields from invoices hosted online.

Use Case Examples

  1. Automatically extract invoice details from a PDF URL to populate accounting software.
  2. Parse multiple invoices by splitting a PDF first and then extracting data from each invoice.
  3. Use webhook callback to receive parsed invoice data asynchronously.

Properties

Name Meaning
Authentication Selects the authentication method to access the PDF parsing service (API Key or OAuth2).
PDF URL The URL of the invoice PDF file to be parsed. The file must be accessible online and contain invoices only.
Custom Fields Comma-separated list of additional fields to extract beyond the default invoice schema.
Advanced Options Additional optional settings such as a webhook URL to receive parsed data asynchronously.

Output

JSON

  • invoiceNumber - Extracted invoice number from the PDF.
  • invoiceDate - Date of the invoice.
  • totalAmount - Total amount due on the invoice.
    customFields - Any additional custom fields extracted as specified by the user.

Dependencies

  • Requires access to an external PDF parsing API service with authentication via API Key or OAuth2.

Troubleshooting

  • Ensure the PDF URL is publicly accessible and contains only invoices; otherwise, parsing may fail or produce incorrect results.
  • If multiple invoices are in one PDF, split them first using a PDF Split module to avoid parsing errors.
  • Check that the authentication credentials (API Key or OAuth2 token) are valid and have necessary permissions.
  • Webhook callback URL must be reachable and correctly configured to receive data if used.

Links

Discussion