Skip to content

Private preview

Form Data Export

Form Data Export turns filled PDF forms into structured files that operations teams can route or reconcile.

Request accessContact supportView capability
ToolboxBrowse the capability directoryWorkflowBrowse pathsAPIRoutes, scopes, and recipesProofInspect evidence

Problem

PDF text can be missing, scanned, inconsistent, or trapped in tables and forms that downstream systems cannot use directly.

Use

Create, inspect, transform, or export text-layer data while preserving warnings and document context.

Evidence plan

What this unlocks

  • Extract all filled form field values as JSON or CSV
  • Run from the document workspace once signed in
  • Available through a scoped private-preview pilot

Workflow

Explore the workflow
Simple pathNIST Cybersecurity Framework 2.0

Cited framework answer

Ask one grounded question, review the cited pages, and save the answer for a compliance note.

  1. 1Upload or select the framework
  2. 2Index text
  3. 3Ask a scoped question
  4. 4Review citations
  5. 5Export the answer

Tool chain

AI ChatKnowledge SearchAI Summarize
Cited answerSource snippetsReviewer note
Explore the workflow
Complex pathAttention Is All You Need

Knowledge and publishing workflow

Extract tables, cluster findings, improve accessibility metadata, convert to archival format, and attach a provenance record.

  1. 1Extract tables and figures
  2. 2Cluster recurring concepts
  3. 3Inspect reading order and metadata
  4. 4Convert to PDF/A
  5. 5Run preflight
  6. 6Export provenance

Tool chain

Extract TablesFinding ClustersAccessibility InspectorPDF to PDF/AStandards PreflightProvenance Ledger
Table exportFinding clustersAccessibility worklist
Explore the workflow

Inputs & outputs

Accepted formats
pdf
Produced formats
csv, json

Common use cases

  • Extract applicant or claim data
  • Reconcile completed forms
  • Feed downstream review or case systems

Workflow steps

  1. 1Select forms or a batch set
  2. 2Map fields and choose the export format
  3. 3Download the structured artifact and retain the source PDFs

Outputs

  • Searchable PDF or text artifact
  • Extraction report
  • Structured data export

Review checks

  • Review OCR or extraction confidence
  • Spot-check page-level warnings
  • Preserve source IDs for traceability

API

Endpoint family: /v1/ocr, /v1/extract, /v1/jobs
POST/v1/rag/searchPOST/v1/ai/chatPOST/v1/processing/extract-tables/{document_id}POST/v1/forensics/findings/clusterPOST/v1/processing/to-pdfa/{document_id}GET/v1/forensics/{document_id}/standards-preflight
  • Export jobs should reference stable document IDs and field names.
  • Use idempotency keys when syncing exports into another system.
Integration overview

Limits

Availability
Private preview
Readiness
beta
Automation
Single-document workflow

Confirm entitlement, input requirements, and deployment fit in the package and API documentation before you automate this capability.

Integration overview

Evidence

Browse proof
Evidence is published with the related workflow and proof artifact when it is available for this capability.

Public example workflows

Integration overview
Simple

Cited framework answer

NIST Cybersecurity Framework 2.0

Ask one grounded question, review the cited pages, and save the answer for a compliance note.

Proof artifacts

Cited answer, Source snippets

Complex

Knowledge and publishing workflow

Attention Is All You Need

Extract tables, cluster findings, improve accessibility metadata, convert to archival format, and attach a provenance record.

Proof artifacts

Table export, PDF/A artifact

Roles and use cases

Operations teams

Operations Automation

Normalize high-volume document streams with repeatable pipelines.

Connected tools

Pipeline Builder, Policy Profiles, OCR

Activity trail: Pipeline run historyAudit record: Batch audit

Research and knowledge teams

AI Knowledge

Chat with document sets using grounded citations and summaries.

Connected tools

AI Chat, AI Summarize, Knowledge Search

Citation evidence: Cited answerGenerated artifact: Summary artifact

Publishing and accessibility teams

Accessibility / Publishing

Remediate, inspect, and validate accessible deliverables.

Connected tools

Accessibility (PDF/UA), Accessibility Inspector, PDF to PDF/A

Generated artifact: Accessibility inspectionVerification result: veraPDF validation result