split-batch

To split a multi-document PDF, you need a tool that detects document boundaries from searchable text and outputs individual PDFs. This Skill extracts page text to find boundary transitions and lets you confirm the split map before generating part files.

Detecting document boundaries from a batch scanned PDF works by applying offline heuristics to OCR-extracted text. It looks for page-number resets, sender changes, or address-block transitions, and can use deterministic separator sheets or consent-gated LLM escalation for ambiguous cases.

Yes, you can split PDFs with pypdf using heuristic document boundaries. This Skill uses pypdf to extract page text and perform the final PDF splitting into part files, relying on offline heuristic detection or separator-sheet mode to find the split points.

Yes, you need to OCR your scanned paperwork before splitting a batch PDF. This is a post-processing tool that requires an already-searchable PDF file to extract page text and detect document boundaries; it does not perform scanning or OCR itself.

When ambiguous document boundaries are detected during paperless workflow splitting, the Skill uses a human-in-the-loop split map for confirmation. If permitted, it can also escalate the ambiguous boundaries to an LLM to reconcile them with the extracted OCR text.

Your multi-document PDF split does not start automatically because the workflow is human-in-the-loop. The Skill proposes boundaries from the OCR text and requires your explicit confirmation, allowing you to merge, split, or edit boundaries before it emits one PDF per document.