Redact
Scrub PII before output / training
Strips personal information out of a job's data before it is delivered, saved, or used to improve the system, so private details do not travel further than they need to.
How it works
Where the PDF redact node works on a document's pages, this node works on the structured data flowing through your job: the names, numbers, and fields the workflow has pulled together. It removes the personal parts before that data moves on.
It matters most at two moments: right before results leave your workspace, and before anything is kept for training or analytics. Redacting first means the sensitive version never gets stored or shared.
Think of it as a privacy checkpoint late in the flow. The workflow can use the full data to do its job, then this node scrubs the personal bits out of whatever is delivered or retained.
Common use cases
Real situations where this node earns its place in a flow.
Anonymize before you keep it
Strip PII from documents you store for review or training so the copy is disconnected from the person.
De-identify before export
Remove personal fields before a document leaves the workflow.
Scrub on the way out
Add a redaction step ahead of any output so nothing sensitive escapes by accident.
Redact in a workflow
A simple example of where this node sits in a flow.
What you set up
- Add it near the end of the flow, before delivery or storage.
- Choose which kinds of personal information should be removed.
- Confirm downstream steps still get the non-personal fields they rely on.
Common questions
How is this different from the PDF redact node?
PDF redact blacks out text on the pages of a document. This node cleans the job's underlying data (the extracted fields and values) before it is delivered or stored.
Does redacting break the rest of my workflow?
It only removes the personal fields you choose. Place it after the steps that need the full data so those run first, then scrub before delivery or storage.
Settings you configure on this node, stored under data.config in the workflow JSON. Leave a field unset to use its default.
kindsComma-separated redaction kinds (ssn, email, phone, routing, account, credit-card, custom); empty means all kinds on.
Type: stringcustomTermsLiteral terms to redact, split on commas and newlines.
Type: stringplaceholderTemplateReplacement text where {kind} becomes the detected redaction kind.
Type: stringDefault:[REDACTED-{kind}]
How this node appears inside a workflow's nodes[] array (the shape you see in the Developer export / import). A full workflow is { nodes: [...], edges: [...] }.
{
"id": "redact",
"type": "card",
"position": {
"x": 160,
"y": 120
},
"data": {
"type": "redact",
"kind": "processor",
"title": "Redact",
"subtitle": "Scrub PII before output / training",
"config": {
"kinds": "",
"customTerms": "",
"placeholderTemplate": "[REDACTED-{kind}]"
}
}
}