PDF

Organise
Merge PDF Split PDF Organise PDF pages Rotate PDF Delete PDF pages Extract PDF pages
Optimise
Compress PDF Flatten PDF Repair PDF
Convert
OCR — make a scan searchable Scan to PDF Images to PDF PDF to images
Edit
Crop PDF Sign PDF Compare two PDFs Add page numbers to PDF Watermark PDF
Secure
Password-protect PDF Unlock PDF Redact PDF Remove PDF metadata

Documents

Document editor Word to PDF PDF to Word PDF to Markdown Excel to PDF

Images

Convert image format Compress image Resize & rotate image

Links

QR code generator URL shortener

Repair PDF

Rebuild a PDF that will not open, and recover what survives.

Repair PDF is a free online tool that rebuilds a damaged or unopenable file around whatever content is still intact, running inside your browser.

File data uploaded this session: 0 B
Add PDF files
Drop PDFs here, or click to choose.

How to repair a damaged PDF

  1. Add the PDF that will not open.
  2. Press Repair PDF. Two approaches are tried, strongest first.
  3. Read the report underneath — it says which approach worked and what that means for the result.
  4. The result says honestly whether the pages came back or came back blank, then offers a Save.

About this tool

Repairing a PDF means rebuilding a file that a reader refuses to open. This free tool attempts it in your browser and tells you honestly which method worked and what was lost.

Most "corrupt" PDFs are not badly damaged at all. The cross-reference table — the index a reader uses to find objects in the file — is easily broken by an interrupted download, a transfer that treated the file as text, or a program that crashed mid-save. The page content is usually sitting there intact behind a broken index. The first thing this tool does is parse the file leniently and write a clean one around whatever it finds, which fixes that whole class of problem with no loss at all: text stays text, images stay images, fonts stay embedded.

When the damage is deeper, it falls back to rendering whatever will still draw and keeping those pages as images. That recovers something from a file that would otherwise be a total loss, but the trade is stated plainly rather than hidden: the text in the result is no longer selectable or searchable. Running the recovered file through OCR will give that back. If neither route finds anything readable, the tool says so instead of handing you an empty document and calling it repaired.

Frequently asked questions

What kinds of damage can be fixed?
Broken cross-reference tables, bad trailers, interrupted saves and files damaged in transfer usually recover completely. Missing or overwritten content cannot be reconstructed by any tool.
Will I lose quality?
Not with the first method, which keeps everything as it was. If the file needs the fallback, pages are recovered as images and the text stops being selectable. The report tells you which happened.
My file was only partly recovered. Why?
Some pages were too damaged to draw. The tool keeps everything it can rather than failing outright, and reports how many pages survived.
Is the damaged file uploaded?
No. Repair runs entirely in your browser, which also means a confidential document that has broken does not have to be handed to a stranger to fix.
The tool could not recover anything. What else can I try?
If you have an earlier copy, a version in an email attachment, or the original source document, those are more likely to work than any repair. A file with substantial data missing cannot be rebuilt.