PDF redact permanently removes marked areas by painting masks into rendered page pixels before building a new PDF.
Every output page is an image. Original text, metadata, attachments, forms and links are not copied; review all pages before sharing.
How to use
- Open a PDF of up to 20 MB and 20 pages.
- Draw rectangles, enter their coordinates, or find keywords, regex or common PII.
- Review every page, confirm the rasterization tradeoff and download the new PDF.
Frequently asked questions
Can the covered text be copied from the output?
No original text layer is included. Masks are painted before images are encoded, so the original pixels under them are not embedded.
Can it find words in scanned pages?
No. Automatic matching reads the PDF text layer, not images. Mark scans manually and check every page.
Why are whole lines sometimes covered?
Automatic matching masks whole PDF text items, which can contain several words. This avoids guessing positions within ligatures or split text.
What is lost when exporting?
All pages are rasterized at up to 144 DPI and 4 megapixels. Search, selection, links, forms, signatures and source metadata are removed.
Pixels, not an overlay
A black rectangle on an original PDF can leave recoverable text underneath. This exporter receives only newly painted page images, never the original PDF objects.
Review and detection limits
Email, phone, SIN and SSN detection uses the existing PII rules. False positives and missed matches remain possible. Regex runs in a disposable worker with a three-second deadline.
