Redact PDF

Actually remove sensitive content — not just cover it with a black rectangle

True redaction is the PDF operation people most often get wrong. The intuitive approach — drop a black rectangle over the sensitive text — looks right but is catastrophically insecure. The text underneath is still in the file. Anyone who copies from the PDF gets the "redacted" content in their clipboard. Anyone who selects the region with the right tool sees the original text. Newspapers have published "redacted" government documents multiple times where the sensitive material was recoverable in seconds. A black rectangle is a drawing, not a redaction.

Real redaction is different. The OkFarsi Redact PDF tool (in development) will permanently remove the underlying text, images, and metadata from the file bytes — not merely cover them visually. The redacted PDF has no hidden text layer under the black marks, no copy-able content in the removed areas, and no recoverable data in the file structure. After download, the file is safe to distribute even under adversarial review.

The tool will support three redaction modes. Visual selection: drag a rectangle over the content you want gone and the tool removes everything inside it. Search-and-redact: type a pattern (an email address, a date, a name) and the tool finds every occurrence and redacts them in one pass — ideal for FOIA-style workflows where you want every instance of a set of identifiers removed. Metadata redaction: strip document metadata (author name, creation-tool signature, revision history) in addition to visible content. All three run as permanent file-level removals, not visual overlays.

How the Redact PDF workflow is being designed

1. Upload the PDF you want to redact

   Drop the file into the redaction workspace. Each page renders as a canvas where you can select the areas or strings to remove.

2. Mark regions for redaction visually

   Drag rectangles over every area that should be removed. The marks are previewed as black boxes, and the underlying text that will be removed is listed in a sidebar for verification — a critical step that stops you from accidentally including non-sensitive content.

3. Or use search-and-redact for pattern-based removal

   Enter an email address, phone number, date, or any identifier. The tool finds every occurrence in the document and marks them for redaction. Review the list, deselect any false positives, then proceed.

4. Strip metadata

   Check the metadata redaction option to remove document properties (author, creation software, revision history, embedded file attachments) — data that survives visual redaction and often leaks more than the text itself.

5. Apply permanent redaction

   Run the job. The tool rebuilds the PDF with the marked regions and metadata genuinely removed from the file bytes, not covered over. The output is safe to distribute even under detailed technical review.

6. Verify the result

   Open the redacted PDF and try to copy from a redacted area. Nothing should copy. Open the file's document properties and confirm the metadata you asked to strip is gone. Send the verified output only after both checks pass.

Planned use cases

• Legal discovery and FOIA releases

  Remove privileged content and personal identifiers from documents released in response to discovery requests or freedom-of-information orders.

• Medical records for research or insurance

  Strip patient names, dates of birth, and other HIPAA-protected identifiers from records before sharing with researchers or third parties.

• Financial documents for due diligence

  Redact account numbers, personal names, and sensitive deal terms from financial records shared during M&A due diligence.

• HR and personnel records

  Remove names and salary details from personnel records before sharing aggregated reporting with external parties.

Privacy & security

Redaction is the most security-sensitive operation in the whole PDF toolkit, and we plan to treat it accordingly. Redaction jobs will run in isolated workers. The tool will not send any redacted content — or the original pre-redacted content — to third-party services at any point. Uploaded files and redacted outputs will be removed from our servers shortly after download. A reminder that no technical control substitutes for careful review: always verify the redacted PDF before distribution, because a missed selection has the same consequence as no redaction at all.

Frequently asked questions