Add your file
Choose or drop a .pdf file. You can paste the contents instead.
PDF Metadata Viewer
Choose a PDF up to 100 MB and the viewer reads its structure in your browser. It shows the header version and any catalog version override, the page count declared by the page tree next to the number of pages actually reached, the Info fields (title, author, subject, keywords, creator, producer, creation and modification dates, with the raw date string kept beside its ISO 8601 reading), the XMP metadata packet, the encryption filter, V and R values and permission bits, linearization, incremental updates with the offset of every cross-reference section, the trailer ID pair, and neutral counts of names such as /JavaScript, /OpenAction, /EmbeddedFile, /AcroForm and /Names. You can copy or download the metadata as JSON and download the XMP packet. If the cross-reference data is damaged, an optional object scan can rebuild it.
File types: .pdfProcessing: This browser
or drop a file here
✓ Stays on this deviceMaximum size: 95 MB
How to use PDF Metadata Viewer
Let it read the file
The file is read and parsed in this browser — nothing is uploaded.
Check the result
Look over what came back, then copy it, download it, or move on.
Want to know how files are handled? Read the privacy notice →
Common questions
How do I view PDF metadata online, such as the author and creation date?
Pick the PDF with the file chooser and this page immediately lists its title, author, producer, creation and modification dates, XMP packet and encryption facts. Nothing has to be installed. The Info table gives every field a status, so a missing field reads Absent instead of looking blank. The Metadata (JSON) panel copies or downloads the versions, page counts, Info fields with raw and ISO dates, XMP status and trailer ID; permissions, structural counts and cross-reference sections stay in their own panels.
Can I check PDF metadata without uploading the file to a server?
Yes: the PDF is parsed by JavaScript inside your own browser tab, and its bytes, Info values and XMP packet are never sent over the network. Nothing is written to localStorage, sessionStorage or IndexedDB either. Once the page has loaded you can go offline and inspect contracts or bids the same way.
How can I see PDF properties on Windows or Mac without installing Acrobat?
Open this page in Edge, Chrome, Firefox or Safari on Windows or macOS and choose the file; author, dates, producer and XMP appear with no Acrobat or other reader installed. File Explorer's Details tab and the macOS Preview inspector show only a handful of fields, whereas this viewer also reports encryption permissions, cross-reference sections and incremental updates.
Does the PDF metadata viewer work on an iPhone or Android phone?
It does: in a current mobile browser you can pick a PDF from Files or Downloads and get the same report you would see on a desktop computer. The whole file is read into memory, so big documents take noticeably longer on a phone, and the 100 MB ceiling is the same on every device.
What should I do when a PDF shows Not a PDF or an unreadable cross-reference table?
Not a PDF means no %PDF- signature exists in the first 1024 bytes, which usually happens when an HTML page, ZIP archive or image was renamed to .pdf. An unreadable cross-reference table normally comes from a truncated download or a broken edit: set Rebuild cross-reference by object scan to On and run again, and the report will say the table was rebuilt and how many objects the scan found.
Why are the author and title of an encrypted PDF shown as unavailable?
In an encrypted PDF the Info strings are stored as ciphertext, and this viewer never accepts a password, so it labels them unavailable rather than printing garbled characters. The Encryption panel still lists the security handler, V, R, key length and each declared permission bit, and the page count is still reported whenever the cross-reference data can be read.
What is the maximum PDF size and which compression can the metadata viewer read?
Files up to 100 MB (100,000,000 bytes) are accepted, and FlateDecode streams with optional PNG predictors are decompressed for xref streams, object streams and XMP. LZW, ASCII85, other filters and the TIFF predictor are reported as unsupported instead of being guessed. The XMP preview shows the first 100,000 characters, and a packet longer than that is not offered as a download.
What is the difference between the PDF Info dictionary and XMP metadata?
The Info dictionary is the older key-value list referenced from the trailer, while XMP is an XML packet attached to the document catalog that can carry richer, namespaced properties. Editing software does not always keep the two in step, so the viewer shows the Info table and the XMP text next to each other; a title that differs between them is how the file was written, not a reading error.
Does /JavaScript or /OpenAction in a PDF mean the file is dangerous?
Not on its own: these counts only record that the names occur in the parsed objects, and forms, print buttons and open-at-page settings use them legitimately. The viewer never runs or analyses scripts and gives no safe or malicious verdict. If the file came from an unknown sender, avoid opening it in a full reader and check it with a security scanner or with the sender.
Does the creation date in PDF metadata prove when the document was made?
No, CreationDate and ModDate are plain text written by the creating software, and they can be wrong, reset or edited by anyone. The viewer shows the raw string next to its ISO 8601 reading with the time-zone offset, and a malformed date keeps its raw text with an empty parsed value. Treat it as a claim and compare it with the XMP dates and the number of incremental updates.
Can I edit or remove the author name from a PDF with this tool?
No, the tool is read-only and never writes a modified PDF, so author, title and XMP cannot be changed or stripped here. It is useful before and after you clean metadata in an editor: run the file again and confirm what the Info dictionary and the XMP packet still contain.
Why is there a PDF page count mismatch between declared pages and pages reached?
The declared number is the /Count stored in the page tree, and the reached number is how many page nodes the viewer actually found by walking that tree; when they differ both are shown and a mismatch is flagged. Damaged or hand-edited files often cause it. Trees deeper than 32 levels, with a loop, or larger than 200,000 nodes are stopped and reported.