DeepRead icon

DeepRead

Process documents with DeepRead — OCR, structured extraction, form fill, PII redaction

Actions3

PII Redact

AI-generated

Summary

Submit a document for PII redaction and poll DeepRead until processing completes or the timeout is reached. On completion, return the API response and, when a redacted file URL is provided, attach the redacted document as binary data.

Inputs

  • Input Binary Field (required) — Name of the incoming binary field containing the document; defaults to data.
  • Language — Document language for PII detection: Arabic, Chinese, English, Hindi, or Spanish. Defaults to English; English is omitted from the API request.
  • Timeout (Seconds) — Maximum time to wait for processing; defaults to 300 seconds. If processing does not complete in time, the operation errors.

Output shape

One output item per input document, with the DeepRead completion response in json and optionally a redacted document in binary.data.

The binary file is downloaded only when the response includes redacted_file_url; its filename is prefixed with redacted_ when the input has a filename, otherwise it is redacted_document.pdf. Failed jobs and timeouts error unless Continue On Fail is enabled, in which case an item with an error field is returned.

Links

Discussion