PDF redaction (true delete)
Mark sensitive regions, create a new PDF, and check the actual output.
Prepare a copy you can inspect
Basic PDFs · up to 50 pages / 25 MiB · local processing
Drag a file here, or click to choose
Processed locally in your browser — never uploadedLinks, annotations, forms, attachments, encryption and complex objects are outside this limited workflow. Unsupported files stop without exporting a partially processed copy.
How it works
Processing stays in your browser. Every output page is rebuilt as an image in a new PDF, with selected regions covered in its pixels. Unboxed areas retain their visible appearance, but every page loses searchable and accessible text. No original pages or resources are copied. The generated file is reopened to check extractable text, supported document Info values, XMP presence and PDF structures. Its actual SHA-256 is calculated.
- 1.Choose your PDF or try the locally generated two-page sample.
- 2.Mark sensitive regions. Optionally enter exact text to check in the output; this does not automatically mark regions.
- 3.Create and check the output. Review page coverage, file hash and any incomplete checks.
- 4.Download and visually review every page. All pages have no searchable text layer.
Checks run on the final file. Exact target text stays in the current session and is excluded from downloaded check records. Missing targets, unchecked content and structure limits are stated separately.
Questions & limits
Which PDFs are supported?
Basic, unencrypted PDFs up to 50 pages and 25 MiB are supported. Files with annotations, internal or external links, forms, attachments, actions or layers are not supported and cannot be exported through this workflow.
Why is every output page no longer searchable?
Every page is rebuilt as an image, including pages without marked regions. Unboxed areas retain their visible appearance, but every page loses searchable and accessible text. This workflow does not run OCR; visually review every image for missed sensitive content.
Do passing checks mean the file is completely safe?
No. Passing applies only to the listed checks and scope. It does not prove that all sensitive information has been removed or cannot be inferred from remaining content.
Related tools
State privacy rights
Deletion laws near this workflow
These source-checked rows connect local redaction and PII cleanup to state privacy deletion rights. The cluster is staged noindex until final legal review.
Updated 2026-07-08 · PrivScrub
Method and limits
PDF output checks apply to supported text, document metadata and structures only. Redacted pages are images; no OCR is run. Visually review the downloaded output and any stated limits before sharing.
Read the processing and verification method