Why use DocSlice

Documents are written for people, not for programs. Every team that has tried to read them with regular expressions or one-off scripts knows how fragile that is: a new layout, a scan instead of a PDF, a different language, and the script breaks. DocSlice replaces that code with one API call.

Typical uses

• Invoices and accounting documents. Read invoice number, date, supplier, totals and line items into your accounting or expense software.
• Contracts and agreements. Pull out parties, start date, term, fees and notice periods, or convert the full text to Markdown for review.
• Product sheets and catalogs. Turn supplier PDFs into the fields your catalog needs.
• Search and AI preparation. Convert documents to clean Markdown, split them by headings and index them, or extract only the facts a prompt needs.
• No-code automations. Call the API from n8n, Make or Zapier to process the attachments that arrive in an inbox or a folder.

Before and after

A supplier invoice arrives as a PDF. Its text looks like this:

INVOICE INV-2847
Date: 2026-05-21
Customer: Acme Srl, Via Roma 1, Milano
Item: 2 x API-PRO licence, 620.00 each
TOTAL 1240.00

With a schema that names the three fields you care about, the answer contains just those fields, with the right types:

{
  "invoice_number": "INV-2847",
  "date": "2026-05-21",
  "total": 1240.0
}

The Quickstart shows how to make this request yourself.

When you do not need it

If your data already arrives structured, you do not need to parse it. An Italian electronic invoice in XML (the SDI format), for example, is already machine readable and can be read directly. DocSlice is for the documents that are not: PDFs, Word files, scans and photos.

Why the location matters

Documents often contain personal and commercial data. DocSlice processes them in Frankfurt, deletes uploaded files right after processing and does not use them for training. That makes it easier to meet your own obligations when you handle other people's documents.