Smart Document Parser icon

Smart Document Parser

Automatically detect and parse documents into structured JSON

Parse Document

AI-generated

Summary

Parse a binary document from workflow input into structured JSON data, automatically detecting or overriding the file type and optionally emitting data per page or row.

Inputs

  • Binary Property (required) — The name of the binary property on the input item that contains the document to parse.
  • Force File Type — An optional override to specify the document type explicitly instead of auto-detection (e.g., PDF, DOCX, XLSX, CSV, JSON, XML, YAML, Image, ZIP, EML, TXT). Defaults to auto-detect.
  • Options — Collection of parsing options including whether to enable OCR for images/scanned PDFs, emit output per page (for PDFs, Word), emit per row (for spreadsheets/CSVs), continue on error, extract metadata only, max extracted text length, and include the original binary data in output.

Output shape

a list of parsed JSON items representing the document contents plus optional attachments, or errors

If 'Emit Per Page' is enabled, output emits one item per page for multipage documents; if 'Emit Per Row' is enabled, one item per row for spreadsheets or CSVs. Attachments extracted from the document are output separately. Errors are output on a dedicated output and processing can continue on failure if enabled. Original binary data can be included in output items.

Discussion