What’s hidden in your PDF?
Before you send a PDF, see what it carries that you can’t see: who made it, earlier versions, attachments, where its photos were taken and text under black boxes. Then clean it in one click.
Choose the PDF to check
or drop it here
Processed in this browser: your file is never uploaded to any server. Turn on airplane mode once the page has loaded and it keeps working. Check it yourself
How to use it
- Choose the PDF. We check it on your device, page by page.
- Read the report: anything marked in amber is something it carries out of sight. Untick whatever you want to keep.
- Tap “Clean and download”. We check the cleaned file again and show you what’s left.
What we check
- Author and software (metadata): the creator’s name, title, subject, dates and the program that made it. PDFs keep these in two places, the Info dictionary and the XMP block, and we read both.
- Text under black boxes: a rectangle drawn on top only covers text from view. If there’s text underneath, we show you what it says.
- Earlier versions: many programs save changes on top of the file and leave the previous state inside, where it can be recovered.
- Photos with camera data: make, model and, when present, the GPS location where they were taken.
- Attachments, hidden layers and JavaScript that runs when the file opens.
What it cleans, and what it doesn’t
- It removes metadata, earlier versions, attachments, JavaScript, and camera and GPS data from photos. Pages don’t change: the text can still be copied.
- It doesn’t delete text under black boxes or hidden layers, because that changes what the page shows. For black boxes, use Redact PDF, which removes the text for real. For layers, flatten the PDF as images.
- If the PDF has a certificate-based digital signature, cleaning it breaks the signature, as any change would.
What we found in testing
We built a test PDF with seven things tucked away, the kind office software and phones leave behind. The report caught all seven in 0.2 seconds:
- 5 details about the author and the software that made it.
- An account number,
4033-1182-7745, under a black box. It could still be copied. - An earlier version saved inside the file, with a different amount.
- A photo carrying the phone model and the GPS coordinates of where it was taken.
- An attached notes file, a hidden layer, and JavaScript set to run on opening.
After cleaning, we checked the file with a different PDF library: no metadata, a single version, no attachments, no JavaScript, and the photo stripped of camera data. The fake redaction and the layer remained, and the report says where to deal with them.
With everyday files: a 5-page PDF exported from a word processor carried 6 author and software details; a 120-page document was checked in full in 2.4 s, and a 15.9 MB scan in 0.2 s.
Limits
- We look for text under black boxes on the first 300 pages, and only under black or very dark rectangles.
- A scan has no text hiding under anything: each page is a picture of the paper. There we check metadata, photos and the rest.
- If the PDF needs a password to open, remove it first with Unlock PDF.
Questions
What is PDF metadata, and why remove it?
It’s information that travels with the file but doesn’t show on the page: who created it, which program, when, and sometimes a draft’s internal title. An ordinary PDF we exported from a word processor carried 6 such details. If the recipient doesn’t need to know them, strip them out.
Isn’t a black box enough to hide something?
No. In our test, an account number covered in black was read back in 0.2 seconds. To make it disappear for real, use Redact PDF, which turns that page into an image with the black painted in.
Does cleaning change how my document looks?
No. Pages stay identical and the text can still be copied; only what you can’t see is removed. That’s why fake redactions and hidden layers aren’t cleaned here.
Is my PDF uploaded to be checked?
No. Checking and cleaning both happen in this browser. Try it in airplane mode, or see Check it yourself.
Other tools
Updated