PDFLean

Redact PDF

Hybrid Processing

Byte-level redaction: matched text removed from PDF content stream with metadata purge; unmatched text stays vector and searchable. Not rasterization.

Mixed processing: local by default; some steps use encrypted cloud with results kept by plan · ≤ 50 MB per fileLocal tools: free & unlimited, no account needed · Cloud tools require a free account (3 tasks per day)

How this tool works (no sign-up needed)

  1. Upload your file above. Each tool shows whether it runs Local (in your browser) or Cloud (encrypted servers).
  2. Adjust options, then click the Run / Convert / Process button.
  3. Wait for progress, then download the result. Use Next tools to continue.

If buttons are not visible, enable JavaScript — the interactive workspace loads in your browser.

pdflean://pdf-redactLOCAL + CLOUD

Free · LocalTrue redaction: matched glyphs are deleted from the content stream

True redaction: matched text is byte-level deleted from page content streams (and from metadata); the rest of the document keeps its vectors and stay searchable. Undo is impossible; keep the original.
Drag files here, or click to browse

Supports application/pdf

Next task:Sanitize PDF →Protect PDF →

Local — your file is processed in your browser and not uploaded to us. No account needed.

What is it?Redact PDF is a PDFLean security & privacy tool: Byte-level redaction: matched text removed from PDF content stream with metadata purge; unmatched text stays vector and searchable. Not rasterization.

TL;DR: Byte-level redaction: matched text is deleted from the PDF content stream with metadata purge and opaque black marks, while unmatched text stays vector and searchable. Everything runs locally in your browser — your file is never uploaded and is cleared when you close the tab; redaction is irreversible (≤10MB), keep a backup.

This tool removes sensitive content from a PDF so it can be shared safely — matched text is deleted at the content-stream byte level, not just covered with a black box. The default local mode runs entirely in your browser with no upload (up to 10 MB); the optional cloud mode requires an account and processes the file server-side with a residue verification report. The output is a redacted PDF, irreversible by design. Key limitation: redaction cannot be undone, so keep a backup of the original file.

About this tool

Permanently redact sensitive keywords from PDF files — pick your mode: LOCAL byte-level redaction in your browser (no upload), or CLOUD true-redaction on our server (pikepdf removes the matched text runs and verifies zero residue). Matched metadata is purged; unmatched text remains vector-searchable. Redaction is irreversible; keep the original. Local mode is free without an account; cloud mode requires an account and counts against your cloud quota (≤10MB).

How to use

1

Upload a PDF

2

Enter keywords to redact (comma/line separated)

3

Pick a mode: LOCAL byte-level redaction (file never leaves the browser) or CLOUD true-redaction (pikepdf removal + residue verification report)

4

Click "Redact PDF"

5

Review hit stats and the residue verification result, then download

FAQ

Will my file be uploaded?

Not in LOCAL mode: everything runs in your browser, the file is never uploaded and is cleared when you close the tab. CLOUD mode requires an account — the file travels over an encrypted channel to our pikepdf engine and is auto-deleted within 24 hours.

LOCAL redaction vs CLOUD true-redaction — which one?

For quick everyday redaction choose LOCAL: ≤10MB, file never leaves the browser, matched text removed at byte level. For legal-grade or highly sensitive documents choose CLOUD: our server-side pikepdf engine rebuilds the content stream to truly remove matched text, runs a residue verification pass (re-reads the output and asserts zero occurrences of the keywords), emits a verification report, and purges matching metadata fields.

Does it redact text inside scanned images?

The current version matches the text layer only. For image-only scans, run Local OCR first to generate a searchable text layer, then redact here; image content is left untouched.

Are there any limits?

Redaction is irreversible — keep a backup of the original; file size is limited to ≤10MB; only matched text glyphs are physically deleted, while unmatched text remains vector and searchable.

Is the protection strong enough?

Everything happens locally in your browser; files never leave your device. Matched keywords are deleted at the byte level from content streams and metadata, covered by opaque black blocks rather than superficial cosmetic boxes.

Can I process multiple files in bulk?

A single tool page handles one task at a time; for bulk work use the Batch Processing workflow and add this tool as a step to run a whole set at once.

Is the processed file still editable?

Redaction is irreversible: matched text is physically deleted from the content stream and cannot be restored. Unmatched portions remain intact vector text and can still be copied or edited. Please keep a backup of your original file.