pdfcosmos

Repair PDF

Rebuild a damaged PDF's structure to make it open again, entirely in your browser.

Free, fast, and privacy-first.

Repair your PDFs

or drag & drop files here

PDF files onlyup to 30 filesup to 100 MB each300 MB total

Repair a corrupt PDF free, in your browser

A PDF that won't open is usually not as lost as it looks. Every PDF carries an index of where its objects live, and that index is the part that breaks most often — after an interrupted download, a file pulled off a failing drive, a sync conflict, or a program that crashed mid-save. The pages themselves are frequently still intact; the file just can't be navigated. Repair PDF rebuilds that structure by scanning for the objects that are still present and writing a fresh, valid file around them.

Nothing is uploaded. That matters more here than on almost any other tool: a damaged PDF is often the only copy of something — a scan of a signed contract, a batch of records, a document whose source is long gone — and the usual advice is to hand exactly that file to an unknown server. Here the recovery runs on your own machine, and your original is never modified; the repaired version arrives as a new download you can compare against it.

It is worth being straight about the limits, because no tool can honestly promise otherwise. This is best-effort recovery. Structural damage is very often fixable. Missing data is not: if bytes were never written, or were overwritten by whatever damaged the file, nothing can reconstruct them. When a file is beyond recovery you get a clear message instead of a plausible-looking but broken download.

There are no settings to choose. Add up to 30 PDFs, press the button, and the rebuild runs locally through a WebAssembly build of qpdf — one file comes back as a repaired PDF, several as a ZIP. It is a sensible first thing to try whenever a PDF refuses to open, opens blank, or is rejected as damaged by another program, since it is quick and cannot make the original any worse.

  • Rebuilds the broken index

    Recovers files whose cross-reference table or object layout was damaged.

  • Your original is untouched

    The repair is written to a new file — nothing overwrites what you still have.

  • Never uploaded

    A damaged file is often the only copy. It stays on your device.

  • Honest about limits

    Missing data can't be invented — an unrecoverable file says so plainly.

How to repair a PDF

  1. Add your PDF(s)

    Drop the damaged file (or up to 30) into the upload area.

  2. Repair

    The tool rebuilds the PDF's structure so it opens again, best-effort, no settings needed.

  3. Download

    Click the button. One file downloads as a PDF, several as a ZIP.

Frequently asked questions

Are my PDF files uploaded to a server?
No. Repair happens entirely in your browser using WebAssembly (qpdf). Your file never leaves your device. You can confirm it in your browser's Network tab: no file data is sent.
What does Repair do?
It rebuilds the PDF's internal structure (the cross-reference table and object layout), which fixes many common problems that make a file fail to open, such as a broken index or minor corruption.
Will it fix every broken PDF?
No. This is best-effort: it recovers many files that other viewers reject, but a severely damaged or truncated file may not be recoverable. If a file can't be read at all, you'll see a clear message.
Does repairing change my content?
No. It only rewrites the file's structure. The pages and their content are preserved as-is.
Which kinds of damage can it actually fix?
The recoverable cases are structural. A PDF keeps an index of where every object lives, and that index is what breaks most often — an interrupted download or upload, a file copied off a failing drive, a sync conflict, or a program that crashed mid-save. Rebuilding the index by scanning the file for the objects that are still there brings a lot of those files back. What cannot be fixed is missing data: if the bytes were never written or have been overwritten, no tool can invent them.
My PDF might be the only copy. Is it safe to use this?
That is exactly why this runs in your browser. A damaged file is often irreplaceable — the only copy of a scan, a contract, or years of records — and most repair sites ask you to upload it to a server you know nothing about. Nothing is uploaded here, so the file never leaves your device. Your original is also left untouched: the repair is written to a new file you download.
Should I try Repair before anything else?
Yes, when a PDF won't open, opens blank, or throws an error about a damaged or malformed file. It is quick, it cannot make things worse (your original is unchanged), and rebuilding the structure is what most other tools try first anyway. If a PDF opens fine but a specific tool rejects it, a repair pass often clears that too.
What if repairing doesn't work?
If the file is too damaged you'll get a clear message rather than a broken download. Worth trying next: any backup or the original source, your email or cloud provider's version history, and — if the PDF still opens in some viewer — printing it to a new PDF from that viewer, which rebuilds the file from what can still be rendered.
Can I repair several PDFs at once?
Yes. Add up to 30 PDFs. One file downloads as a repaired PDF, and two or more download together as a single ZIP.
Is Repair PDF free?
Yes, it's completely free to use, with no sign-up and nothing to install.