Repair PDF
Open a damaged file
A PDF that will not open is usually intact apart from the index that says where everything is. This rebuilds that index from the objects still in the file, tells you what it found, and only hands back a document it has reopened and read successfully.
Questions
- Are my files uploaded anywhere?
- No. Every tool on this site runs entirely inside your browser using JavaScript — your file is never uploaded to a server, because there is no server. You can confirm it by opening your browser’s network tab, or by disconnecting from the internet once the page has loaded: the tool still works.
- Will this fix any PDF?
- No, and it will tell you when it cannot. It rebuilds a file’s structure — the cross-reference table, the offsets, the trailer, stream lengths — from the objects that are actually there. What it cannot do is invent bytes that were never in the file. If a download or a transfer stopped part-way, the missing content simply is not present, and no tool anywhere, on a server or otherwise, can rebuild it. Truncation is detected and reported as exactly that.
- What kinds of damage can it fix?
- The common ones: a cross-reference table whose offsets no longer match the objects, a missing or misdirected pointer to that table, a trailer that names no document catalog, junk in front of the file header that shifts every offset, and streams declaring a length that is not their real length.
- It says nothing is wrong with my file — now what?
- Then the file’s structure is sound and it is handed back untouched rather than resaved. If it still will not open, the problem is elsewhere: the wrong file extension, a password the reader is not asking for, or an application that cannot handle something the document uses.
- Can it open a password-protected PDF?
- No. Repairing the structure does not decrypt anything — the file stays protected and still needs the same password afterwards. This tool cannot remove or recover a PDF password.
- Some pages came back blank — why?
- Those pages’ contents could not be decoded, usually because the damage reached the page data itself rather than only the index. They are listed by page number so you know what is missing. The rest of the document is genuinely recovered — a partial rescue, honestly labelled.