File Metadata Viewer
Size, type, timestamps and signatures
Inspect any file: its real type from the magic bytes, its timestamps, its SHA-256, and whether the extension is telling the truth.
File Metadata Viewer
Runs in your browser · nothing uploaded01 — How to use File Metadata Viewer
02 — Getting good results
03 — Questions
No. The browser reads it from your disk into memory in this tab. Detection reads the first thirty-two bytes; hashing reads the whole file locally with the browser's own crypto implementation. Nothing is transmitted.
Because the leading bytes do not match any signature the tool knows. Plain text, CSV, JSON, source code and many other formats have no signature at all — they are just characters, with nothing at the start to identify them. An unrecognised result is common and usually means exactly that.
It is a fingerprint. Two files with the same SHA-256 are, for every practical purpose, the same file; a single changed byte produces a completely different value. It is what you compare against a published checksum to confirm a download arrived intact, and how you check whether two copies have diverged.
No, and you should not treat it as if it could. It reports what a file appears to be, which is useful context — an attachment named invoice.pdf that is really a Windows executable is a strong signal something is wrong. But a correctly named file can still be malicious, and this tool does not scan for that.
04 — More about File Metadata Viewer
Magic numbers and how type detection works
Most binary formats begin with a fixed byte sequence that identifies them, known as a magic number. A PDF starts with the ASCII characters %PDF. A PNG begins with a byte of 0x89 followed by the letters PNG and a carefully chosen sequence of line-ending bytes, designed so that a file corrupted by a text-mode transfer becomes detectably invalid. A JPEG starts 0xFF 0xD8 0xFF. These markers exist precisely because filenames are unreliable: the same file arrives with different names on different systems, and the format needs to be identifiable from its contents alone. Reading them is how operating systems, web servers and command-line tools decide what a file really is.
Containers, and why one signature can mean several things
Some signatures identify a container rather than a format. The ZIP signature is the clearest case: a ZIP archive is a general-purpose bundle, and a great deal of modern software stores its documents as one. A .docx, .xlsx, .pptx, .odt, .epub, .jar and .apk are all ZIP archives that differ only in what they contain and how the contents are named. Distinguishing them requires unpacking the archive and looking inside, which is a substantially different operation from reading a header. The same applies to the ISO base media format, where AVIF images, HEIC photographs and MP4 videos share an identical outer structure and differ in a four-character brand code a few bytes in. This tool checks those brands specifically, which is why a HEIC photograph is not reported as a video.
Why a mismatch matters
A file whose name and contents disagree is not automatically sinister — renaming happens accidentally all the time, and some systems strip or replace extensions as a matter of course. But it is always worth knowing about. In the benign case it explains why something will not open: an application asked to handle a .jpg that contains PDF data will simply fail, often with an unhelpful message, and the cause is invisible without looking at the bytes. In the less benign case, a deliberately misleading extension is a long-standing technique for getting a user to open something they would not have opened knowingly. Either way, the disagreement itself is the useful signal, which is why it is reported first rather than buried among the other fields.
61 more tools
where this came from.
Same shell, same shortcuts, same promise: nothing leaves your browser.