Skip to main content
Human review lets you add a verification step to your workflow where a person reviews and optionally corrects AI-extracted data before it proceeds.

​
When to use human review

  • High-stakes documents: financial, legal, or compliance documents where accuracy is critical
  • Low-confidence extractions: only review documents where the AI is uncertain
  • Training period: review all documents initially, then narrow to the runs that need attention as you refine your schema

​
Adding a review step

  1. Open your workflow in the editor
  2. Add a Review action node after your extract action
  3. Connect: Extract Action → Review Action → (next step)
  4. Configure the review node

​
Configuration

The node has no trigger setting of its own: a run pauses whenever it reaches the node. Decide which runs reach it with a condition edge on the incoming connection.

​
How the review workflow works

  1. The run reaches the review action and pauses
  2. A review task appears in the review queue
  3. A reviewer claims and reviews the extracted data
  4. On approve: the run resumes with the (possibly corrected) data
  5. On reject: the run is marked as failed

​
Example workflow with review

A common pattern is to extract data, review it, then send it to a callback: Upload Trigger → Extract Action → Review Action → Callback Output This ensures a human validates every extraction before results are delivered.

​
Tips

Start by connecting Review directly so every run pauses. Once you trust the extraction, turn on the extract node’s Review advice and route only {{metadata.reviewAdvice.needsReview}} is true into review.
  • Write clear, specific review instructions to guide reviewers
  • Route on the signals that matter to you: Review advice, unverified fields, contested bindings, validation failures
  • Review tasks are visible to all users with review permissions in your organization