Repair a corrupted ZIP and extract what's inside — no upload

A broken archive is rarely all-or-nothing. Most of the files inside usually survived; the job is getting them out.

Your file never leaves your device — repair runs in your browser. 0 bytes uploaded.

Your data is the big block — usually intact. What breaks is the small index. Repair rebuilds it.

Drop a file here or browseTap to choose a file

ZIP · Office · PDF · video · JPG · PNG · RAR/7z · SQLite — repaired right here in your browser. Nothing is uploaded

0 bytes uploadedFree: 3 repairs/day — up to 500 MB video, 100 MB docs & archives, 50 MB photos. Free download. Single repair €5.90. No account required.

In short

This tool repairs corrupted and truncated ZIP archives: it rebuilds the missing table of contents and extracts every file whose data survived, even when one strict unzipper refuses the whole thing. It all runs in your browser — the archive is never uploaded, so 0 bytes leave your device, which you can confirm yourself in the browser's Network tab. The one honest limit: if a file's compressed data is itself damaged, that entry can't be recovered.

ZIP won't open or extract? Repair it now

  1. Drop the archive
  2. Repair and extraction run locally
  3. Download your files

To fix a corrupted ZIP, you rebuild the archive's table of contents and pull out every file whose data survived. IntactFile does exactly that, in your browser: drop the archive, the engine scans it locally, reconstructs the structure, and hands back the salvageable files, even when the archive as a whole is past saving. Nothing is installed, and the archive is never uploaded anywhere.

How ZIP corruption works

A ZIP is a row of independently compressed files followed by a central directory: the archive's table of contents, stored at the very end. Extractors read that directory first to learn what's inside and where. That layout explains almost every broken archive you'll meet:

When a download or copy is cut short, the end of the file is what goes missing, which means the table of contents is lost while the files themselves sit intact in the earlier part of the archive. Your unzipper refuses the whole thing because its index is gone, not because your files are.

The same layout makes partial salvage possible: ZIP writes a small local header in front of every file it stores. By scanning for those headers, the engine finds each surviving entry and extracts it individually, no table of contents required. So the honest promise is not "fixed or failed" but every file whose data survived comes out.

The errors this covers

Different tools word the same damage differently. If you're seeing one of these, you're in the right place:

Unexpected end of archive

WinRAR — the file ends before the table of contents. Classic truncated download.

The archive is either in unknown format or damaged

WinRAR — the structure can't be parsed at all; the start of the file may be damaged too.

Cannot open file as archive

7-Zip — no recognizable archive structure where it expects one.

End-of-central-directory signature not found. Either this file is not a zipfile, or it constitutes one disk of a multi-part archive.

unzip on the command line — the most literal of the four: the end record that anchors the table of contents is missing.

Windows Explorer has its own wording: if you're staring at The Compressed (zipped) Folder is invalid, there's a dedicated page for that error, including the free fixes to try first.

Why archives break

Three causes account for nearly all of it. Interrupted transfers: a download that stalled at 96% or a USB stick pulled mid-copy: the tail of the file, table of contents included, never arrived. Email and cloud mangling: gateways and sync clients occasionally truncate or re-encode attachments, and large multi-file cloud downloads are a known weak point. Storage trouble: bad sectors corrupt bytes in place, the only cause of the three that can hit the compressed data itself.

What this can and can't fix

Can fix

  • Missing or damaged central directory — rebuilt from the local headers
  • Truncated archives: every complete entry before the cut is extracted
  • Archives one strict extractor refuses but whose entries are intact
  • Partial salvage as a first-class result — you get a per-file report of what came out

Can't fix

  • Entries whose compressed data is itself damaged — DEFLATE carries no redundancy, so a corrupted stream can't be decompressed past the damage
  • Password-protected and AES-encrypted archives repair structurally, but the contents still need the password
  • Multi-part archives with a missing volume — every part must be present
  • Files that read as 0 bytes: that's a data-recovery problem, not a repair problem

If nothing can be salvaged, the tool says so and explains why, and you're never charged for a failed repair.

FAQ

How do I fix a corrupted ZIP file?

Drop it into the tool above. The repair runs locally in your browser (the archive is never uploaded), and the engine extracts every file whose data survived, even when the archive as a whole can't be made valid again. The free tier covers archives up to 100 MB, 3 repairs per day, with a free download and no account.

What does 'unexpected end of archive' mean?

It's WinRAR's way of saying the file stops before the archive's table of contents, which lives at the very end of a ZIP: the classic sign of a truncated download or copy. The files stored before the cut are usually intact; the archive just lost the index that points to them, and that index can be rebuilt.

Can I extract files from a damaged ZIP?

Very often, yes, partially. ZIP stores a small header in front of every file it contains, so even when the archive's central index is gone, each surviving entry can be found and decompressed on its own. Getting 47 of 50 files out is a normal result.

Why does 7-Zip say cannot open file as archive?

7-Zip couldn't find a structure it recognizes: the ZIP signature or the end-of-archive record is missing or damaged. Common causes: a truncated file, overwritten first bytes, or a file that isn't really a ZIP (a failed download page saved with a .zip name). A local-header scan settles which one it is.

Does re-downloading fix a corrupted ZIP?

Yes, frequently — and always try it first. If the archive came from the web, email, or cloud storage, the server's copy is usually fine and the damage happened in transit. Repair is for when re-downloading isn't possible: the source is gone, or the archive is the only copy.

Related: Windows won't open the folder — The Compressed (zipped) Folder is invalid · Excel file won't open? An .xlsx is secretly a ZIP — repair Excel files ·verify the zero-upload claim in your Network tab