Protect PII while preserving supported document/file workflows instead of extracting everything into plain text first.
text · json · pdf · docx · xlsx · csv · image · email · dicom · transcript
Text, JSON, PDF, DOCX, XLSX, CSV, image, email, DICOM and transcript workflows are exposed by the engine.
PDF redaction can use OCR where required, with configurable page limits and runtime dependency checks.
Image redaction uses OCR language selection and automatic script detection when Tesseract support is installed.
DOCX processing walks document containers; XLSX/CSV processing can use structured policies for columns/sheets.
Email processing can protect message content and can process supported attachments, reporting partial success if an attachment fails.
DICOM tag handling masks direct patient/physician fields and text-like tags; transcript/SRT/VTT processing preserves timing lines while protecting text.