IntactFile

Repair a corrupted Word document — without your file leaving your device

A .docx that won't open is usually a broken wrapper around text that survived. Rebuild the wrapper; keep the words. Nothing to install.

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.

IntactFile repairs corrupted or unreadable .docxdocuments by opening them at the archive level and salvaging the text, styles, and images that survived. It runs entirely in your browser — your document is never uploaded, and you can confirm 0 bytes leave your device in the Network tab. The honest limit: text whose underlying data was truly destroyed can't be brought back, only the parts that are still readable.

Repair your document now

  1. Drop the .docx
  2. Repair runs locally
  3. Download the rebuilt document

To repair a corrupted Word document, you open it at the archive level (a .docx is a ZIP of XML parts), salvage the readable parts, and rebuild a valid package from them. IntactFile does that in your browser: drop the file, the engine reads it locally, and you download the result with a report of what came back. Your document is never uploaded.

Try Word's own repair first

Word ships a repair mode most people never notice, and it's free:File → Open → Browse, select the damaged file, then click the small arrow next to the Open button and chooseOpen and Repair. When Word can route around the damage, you're done. When it can't — it reports the file is corrupt, opens a blank page, or throws Word found unreadable content in name.docx and then recovers almost nothing — the package is damaged in a way Word's own path can't navigate. That's the case for archive-level repair, and to see why it works you need one fact about what a .docx actually is.

A .docx file is secretly a ZIP

Rename any healthy document from letter.docx toletter.zip and open it: you'll find folders of XML files.word/document.xml holds your actual text and paragraphs; other parts hold styles, fonts, and any embedded images, plus a[Content_Types].xml manifest and _relsrelationships that tie them together. Because it's a ZIP, atable of contents at the end of the file records where each part lives.

So "corrupted Word file" means one of two things: the ZIP container is broken (an interrupted save to a network drive or USB stick, a truncated email download — the table of contents is damaged while the text part survives), or one XML part inside is malformed while the rest are fine. Both are partial-damage situations, and that's exactly why ZIP-level repair recovers documents: the engine finds each surviving part by its own local header, no table of contents needed, and rebuilds a valid package from what's readable — the same mechanism asZIP repair, pointed at Word.

What gets recovered

Setting expectations precisely: body text, paragraphs, and styles recover best, because they live indocument.xml and the style part that archive-level salvage targets. Embedded images come back when their media parts survived. More fragile are tracked changes, complex fields, custom XML, embedded objects (OLE), and VBA macros in a.docm — they live in separate parts, and whichever of those were damaged or cut off don't return. The result report tells you what was rebuilt, so you know the gaps before you open the file.

What this can and can't fix

Can fix

  • Broken ZIP container around an intact document.xml — the common case
  • Truncated documents: every readable part before the cut is salvaged
  • Files Word's Open and Repair rejects or opens blank
  • A malformed XML part while the rest of the package is fine

Can't fix

  • A document.xml whose compressed data is itself destroyed — that text no longer exists in the file
  • Password-encrypted documents without the password
  • Old binary .doc (OLE) files — a different format entirely
  • Files that read 0 bytes — a data-recovery problem, not a repair problem

If the salvage comes back empty, we tell you why, and a failed repair is never charged.

FAQ

How do I repair a corrupted Word document?

Two steps, in order. First try Word's built-in Open and Repair: File → Open → Browse, select the file, then click the small arrow next to the Open button and choose Open and Repair. If that fails or opens a blank document, drop the .docx into the tool above: it opens the file at the archive level, salvages the readable parts, and rebuilds a valid document from what survived — all in your browser, never uploaded.

Word says it 'found unreadable content' in my .docx — what do I do?

The full message is "Word found unreadable content in name.docx. Do you want to recover the contents of this document?" Click Yes and let Word try first; often that's enough. If recovery fails or returns a document missing most of its text, stop re-saving over the file (each save can overwrite what's left) and run a structural repair on a copy. That message means the package parsed but at least one XML part inside is malformed — exactly the case archive-level salvage is built for.

Can I recover text from a .docx without Microsoft Word?

Yes. A .docx is a ZIP archive of XML files, and the main text lives in a part called word/document.xml. That part can be read with archive-level tools — no Word, no Office license required. The tool on this page does exactly that in your browser and rebuilds a document from the parts that survived.

Why does my Word document open blank or with strange characters?

A blank document usually means the package opened but the part holding your text (word/document.xml) is damaged or unreachable. Strings of random characters, or a file that opens in Notepad as garbled binary, usually mean you're looking at the compressed bytes directly — the ZIP wrapper is intact but the XML inside is corrupt. Archive-level salvage can often still reach the readable parts because it doesn't rely on the package's own wiring being intact.

What's the difference between .doc and .docx for repair?

They're completely different formats. A modern .docx (Word 2007 and later) is an Open XML package — a ZIP of XML parts — which is what this tool repairs. An old .doc is a binary OLE compound file with a different internal structure; if your file is really a .doc, converting or repairing it follows a different path. Check the extension: this page targets .docx (and .docm macro documents, which are the same package format).

Is it safe to upload confidential documents to an online repair tool?

Contracts, NDAs, medical letters, legal filings — a broken Word file is often the last thing you'd want on a stranger's server, on their retention terms. Most "online" repair tools require exactly that upload. Here there is no upload at all: the repair runs in your browser, the file is read from your disk and rebuilt in your tab's memory, and you can confirm 0 bytes leave your device in the Network tab.

Related: the exact recovery prompt — Word found unreadable content · spreadsheet won't open — repair Excel files · any archive that won't open — repair ZIP files · contracts and records — repair PDF files ·verify the zero-upload claim