• Custom n8n node
  • Documents & AI
  • AI-agent tool

PDF

Split PDFs by page ranges and count pages, entirely inside the workflow.

How it works in practice

One big scan, cut into one file per document

How we apply it

It is the cutting step behind our document robots: the AI says where each document starts, and this step makes the split, for example for scanned invoice batches of an accounting firm.

A small, focused PDF toolkit with two operations. Get Page Count returns the number of pages of a PDF. Split PDF cuts a PDF into separate documents from a range expression such as 1,2,3-5,8 and returns one binary PDF per range, named by its pages. Page numbers are validated against the document, so a bad range fails with a clear error.

We needed the second half of the Document AI splitter pipeline: the splitter says which pages belong to which document, and something has to actually cut the file. The node runs locally with no external API and stores output through n8n's binary-data helper, so large files do not bloat execution data and filesystem or S3 binary storage modes keep working.

It is used in 8 production workflows, mostly for page counts before sending a file to a paid OCR service, and for splitting scanned invoice or statement batches into one file per document before archiving them in SharePoint or object storage.

Limits: that is all it does. There is no merge, rotate, text extraction or form filling.

For your technical team: it installs from npm as

n8n-nodes-kt-pdf

What it means for you

  • No manual splitting in a PDF viewer

  • Fewer wasted pages sent to paid reading services

  • Nothing extra to buy or connect

Want PDF in your n8n?

Tell us what the workflow has to do and what it connects to. We will say whether this node covers it as it stands, what we would add, and when a different approach is the better one.