PDF Redactly is private redaction for PDFs. Detect and hide sensitive information without your documents ever touching a server.
Detection runs entirely in your browser — no cloud AI, no LLMs, no servers ever see your file.
Everything happens locally in your browser. You stay in control of your data.
Watch demoDrag and drop any PDF into the workspace.
Our on-device AI scans every page for 30+ categories.
See results, approve, reject, or add your own redactions.
Download a clean, properly redacted PDF. All local.
Use cases
Pick your industry profile and detection adapts. The patient-name rule on a medical letter is different from the client-name rule on a legal engagement — PDF Redactly knows the difference.
Legal
Attorneys, paralegals, legal ops
Medical
Clinics, billing teams, research coordinators
HR
People ops, recruiting, internal investigations
Personal
Anyone sending a PDF that has more on it than they want shared
Detection coverage
Not a generic catch-all. The medical profile knows what an MRN looks like; the legal profile knows what a case number looks like; the personal profile catches the bank-statement essentials. Pick yours and the rules adapt.
Names and identifiers across formats
Issued by agencies
Reach-me-here data
Account-level identifiers
Clinical document identifiers
Filings, contracts, briefs
Plus your own custom phrases — typed in, clicked from a doc, or dragged out of an existing redaction. Everything stays on this device.
New
The engine is a starting point, not the last word. Add your own phrases, click any word in the page to mark it, or drag the edge of an existing redaction to extend or shrink it.
Type a phrase and it's redacted everywhere it appears in the document — case-insensitive substring match. Save to use on every future document, or scope to just this one.
Hover the page → cursor turns to a crosshair → click a word. PDF Redactly adds it to your custom list and redacts every occurrence instantly.
Grab the left or right edge of any redaction box. Drag out to include adjacent words. Drag in to keep only what you want covered. The change applies everywhere the phrase appears.
How we differ
Most tools say “private” but still upload your file. PDF Redactly is the only one where the file genuinely never leaves your browser. The differences compound from there.
| Feature | PDF Redactly | Other PDF redaction tools |
|---|---|---|
| Where your file is processed | 100% in your browser. Never leaves your device. | Encrypted upload, decrypted server-side, or 'hybrid' inference. |
| Cloud AI / LLM dependency | None. On-device entity recognition + deterministic rules. | Server-side LLMs and ML pipelines. |
| Industry-tuned profiles | Personal, Legal, Medical, HR — pick one and the rules adapt. | Generic detection across all document types. |
| Custom redaction terms | Add your own phrases. Save globally or scope to one document. | Limited or absent. |
| Click-to-mark + drag-adjust UX | Click any word; drag any redaction edge to extend or shrink. | Manual draw boxes only. |
| Pricing transparency | Visible: $0 / $2.99-mo / $9.99 lifetime / Business custom. | Hidden behind a click or sales call. |
| Watermark on free output | Diagonal stamp + footer (so you know what you got for free). | Often unannounced. |
| Source you can audit | Open about every detection layer. Privacy claim is verifiable. | Closed engine. Privacy claim is a promise. |
Compliance posture
PDF Redactly doesn’t make HIPAA / GDPR / SOC 2 claims with certificates. It does something better: removes itself from the data flow entirely. The frameworks all start with “does the vendor receive personal data?” — for PDF Redactly, the answer is no.
Patient data never leaves your device, so there's no covered-entity transfer to worry about. Your BAA list stays short.
We don't collect what we don't need. The only personal data on our servers is your email + Stripe customer ID. No document content, ever.
Because the file path doesn't traverse our infrastructure, our SOC 2 surface area is tiny. Not a substitute for your own audit, but a much shorter conversation.
After the first page load, redaction works without an internet connection. Useful for lawyers reviewing a deposition on a plane or clinicians on a hospital VPN.
We don’t hold formal certifications today. The architecture stands on its own — read the source, run our regression suite, or read our privacy policy for the long version.
Privacy promise
We don’t pipe your documents through an LLM in the cloud, because we don’t pipe them anywhere. Detection runs in your browser tab. Your file never leaves your device.
Parsing, detection, and redaction happen in this browser tab. We never receive the document.
No API calls to OpenAI, Anthropic, Google, or any third-party model. Nothing to leak.
Pattern matching plus a small on-device entity model. Documented, auditable, no black box.
PDF REDACTLY is private by design. No cloud.
No tracking. No logs. Just powerful redaction that stays on your device.
Runs in your browser using WebAssembly.
We never see your files. Ever.
No analytics. No cookies. No fingerprinting.
Transparent, auditable and verifiable.
All plans include local processing, all features, and all document types.
30-day money-back guarantee.
Detects names, IDs, financials, medical info, and more.
Runs entirely in your browser. No data leaves your device.
Verify, export logs, and inspect every redaction.
Correct. There is no upload endpoint. Open your browser's network tab while you scan — you'll see zero requests for your file. The model and redaction engine ship as JS/WASM and run locally.
Permanent. Underlying text and image data inside the redacted region are destroyed in the output PDF — not just visually covered. Copy-paste reveals nothing.
30+ out of the box, including names, addresses, phone numbers, emails, SSNs, credit cards, bank account numbers, dates of birth, medical record numbers, salary figures, case numbers, and more.
Acrobat is $19.99/mo and uploads your file to Adobe's cloud for many features. PDF Redactly is $2.99/mo (or $9.99 once for lifetime), fully local, and includes auto-detection that Acrobat charges extra for.
We ship Legal, Medical, and HR packs on Pro. Each adds detection patterns specific to that industry (e.g. ICD-10 codes, case numbers, performance review identifiers).
Yes. The browser app works on mobile Safari and Chrome. The on-device model is small enough to run smoothly on modern phones.