Skip to main content
The following information applies only to dedicated instance and in-VPC deployments of Unstructured Business.Contact your Unstructured account team or open a ticket in the Unstructured Support Portal to have it enabled for your deployment.
Type: pii_redactor Subtype: pii-guard
PII detection is probabilistic and is not a guarantee. Do not use this node as your sole privacy or compliance mechanism. For details, see Limitations.
For more information, see PII redaction.

​
Workflow placement

  • Place the PII Redaction node anywhere after the Partitioner node and before the destination.
  • The recommended placement is immediately after the Partitioner node, so that PII is redacted before any downstream Chunker or Enhancement node processes the text.
  • If the Partitioner node uses a VLM-based strategy, the original, unredacted document content is still sent to the external VLM provider during partitioning, before this node runs.

​
Settings

​
string
default:"redact_with_audit"
What the node does with detected PII. Allowed values:
  • detect_only: Leave text unchanged; only record findings in metadata.pii_detections.
  • redact_with_audit: Anonymize the text and record findings in metadata.pii_detections.
  • redact_only: Anonymize the text; write no metadata.
Learn more about redaction modes.
​
string
default:"replace"
How each detected PII span is anonymized when mode redacts text. Allowed values: replace, redact, mask, hash. Learn more about anonymization methods.
​
number
default:"0.4"
Minimum detection confidence, from 0 to 1, for a finding to be acted on. Learn more about the confidence threshold.
​
string
default:"*"
Single character used to overwrite PII when operator is mask. No effect for any other operator value.
​
string
Secret key for the hash operator. Tokens are HMAC-SHA256 keyed with this value. Required when operator is hash and mode redacts text (redact_with_audit or redact_only), unless the deployment sets a default through the PII_HASH_SECRET environment variable; a per-node hash_secret value takes precedence over the environment variable. Learn more about the hash secret.