DocuPipe icon

DocuPipe

Automate document processing, data extraction, and classification with DocuPipe

Actions10

Extraction → Extract

AI-generated

Summary

Extract structured data from a fully processed document using a specified schema and extraction engine version.

Inputs

  • Document ID (required) — The ID of the document to extract data from; must be in completed status.
  • Schema (required) — The schema defining extracted fields, selectable from a searchable list or by ID.
  • API Version (required) — Select the extraction engine version (V3 recommended or V2 legacy).
  • Additional Fields (V3) — Optional parameters for V3: Effort Level (only standard currently), and extraction guidelines text.
  • Advanced Options (V3) — Optional V3 advanced options: pages to extract (0-indexed comma-separated), timeout seconds, and useMetadata flag.
  • Additional Fields (V2) — Optional parameters for V2: display mode, effort level, extraction guidelines, and document split mode.
  • Advanced Options (V2) — Optional V2 advanced options: pages to extract (0-indexed comma-separated), timeout seconds, and useMetadata flag.

Output shape

a single JSON object representing the extraction result

The output JSON contains the extracted structured data as defined by the provided schema. Use the API version to control extraction engine behavior and features. The document must be fully processed (status completed) prior to extraction.

Links

Discussion