PDF Metadata Editor

Read the title, author, subject, keywords and creator a PDF stores, then export a copy with the values you set.

Files stay on your device No sign-up Free to use
How this works

The tool runs in this browser. Your file or text is not uploaded to UseFreeTools. Check this tool's limits for anything it may save on your device.

Privacy details

Read or write controls

Drop your PDF here

or choose one from your device

One PDF file, up to 50 MiB. The file is not uploaded.

    Used by the write mode. Leave empty to remove the entry when removal is on.

    Used by the write mode. Leave empty to remove the entry when removal is on.

    Used by the write mode. Leave empty to remove the entry when removal is on.

    Used by the write mode. Separating terms with a comma keeps the entry readable.

    Used by the write mode. This names the application that made the file, not the PDF producer.

    On: an empty information field or XMP property is taken out of the document. Off: anything you leave empty keeps the value it already had.

    XMP is a separate XML record that repeats some of the same names. It is read and reported either way.

    Used while XMP metadata is set to write. Empty removes the property when removal is on.

    The first value of the XMP creator list. Used while XMP metadata is set to write.

    Used while XMP metadata is set to write. Empty removes the property when removal is on.

    Used while XMP metadata is set to write. Separating terms with a comma keeps the entry readable.

    A short copyright statement. Only write a claim you hold.

    How to use PDF Metadata Editor

    1. Add one PDF in the PDF file field. The file is read in this browser and never uploaded.
    2. Leave What to do on Read the fields already stored, then run once to see the information entries and the XMP properties the file carries.
    3. Switch to Write the fields below and type the entries you want. A field you leave empty is removed while removal is on.
    4. Set XMP metadata to write the five named XMP properties, or to remove the whole packet, and leave it alone when the XML record should not be touched.
    5. Select Read or write and download the new copy.

    Example: PDF Metadata Editor

    Read both metadata records of one report, then rename the title in each.

    You add
    A three-page PDF titled Quarterly report with the author J. Field and an XMP packet that repeats the same title. First run with What to do set to Read the fields already stored. Then run again with Write the fields below, Title set to Quarterly report 2026, Author left empty, and XMP metadata set to Write the XMP properties below with XMP title set to Quarterly report 2026.
    You get
    The read run lists the five information entries, the dates, and every XMP property it could read. The write run reports how many information entries and how many XMP properties changed, and the downloaded copy shows the new title in both records with the author entry gone.

    Options

    What to do
    Read the fields already stored returns a report and changes nothing. Write the fields below saves a new file with the values you typed.
    Remove the fields I leave empty
    On, an empty box takes that entry out of the document and out of the XMP record. Off, the value the file already had is kept, which is the safer choice when you only want to add a title.
    XMP metadata
    Leave the packet as it is changes nothing in the XML record. Write the properties below replaces the five named properties and copies the rest of the packet through, including namespaces this tool does not understand. Remove the whole packet takes the record out of the file.
    Keywords
    One text entry, not a list. Most readers show it as written, so separating terms with a comma keeps it readable.

    Supported inputs and limits

    File sizeOne PDF up to 100 MiB
    Entries editedFive info entries and five XMP properties
    Left aloneProducer, dates, page content and unseen app data
    One PDF per run, up to 100 MiB, and every page is kept. Five information entries are editable, each capped at 300 characters, plus five XMP properties capped at 500 characters each. An XMP packet above 256 KB is reported but not rewritten, and it can still be removed as a whole; a compressed packet is expanded with the browser's own streaming decoder inside that same bound, so a forged size cannot allocate more than the limit. A packet this tool cannot scan, one stored with another filter or a predictor, one that names a property twice, and one whose property holds nested elements are all reported and left alone rather than half-rewritten. The producer, creation date, modification date, bookmarks, annotations, page content and attachments are never touched, and XMP properties this form does not name are copied through with their namespaces intact, matched by namespace URI rather than by prefix. Clearing these records is not a promise of anonymity, because page content, embedded files and other containers can still name the author. A signed PDF downloads without a valid signature because the bytes were rewritten. Password-protected files are refused.

    Where your input is processed

    This tool processes your input in this browser. Your text and files are not uploaded to UseFreeTools. Check this tool's limits for anything it may save on your device.

    One file can carry several kinds of metadata

    The information dictionary holds short text entries next to the document catalog: a title, an author, a subject, a keyword string, the creating application and the producing library. An XMP packet is a separate XML record, usually stored as a metadata stream on the catalog, and it can repeat the same names in another form. Some files carry EXIF or IPTC data inside embedded images as well. Editing the dictionary and the named XMP properties is the part this tool can verify, which is why the report states plainly what it changed and what it did not.

    How the XMP packet is rewritten

    The packet is edited as text rather than rebuilt. The reader finds the description element, replaces only the value of a named property, and leaves every other byte of the packet where it was, so an unfamiliar namespace, an unknown attribute or a nested structure survives the round trip. When a value has to be added, the standard namespace declaration is added first and the packet keeps the prefix it already used. A packet above the reading limit is refused for rewriting instead of being cut down.

    Questions about PDF Metadata Editor

    Does clearing the author make the document anonymous?

    No. This tool edits the information dictionary and the five named XMP properties. An XMP property it does not name, an attachment, a comment or the wording of the page itself can still identify a person, so treat the export as one step rather than a guarantee.

    What happens to XMP properties this form does not name?

    They are copied through the rewrite exactly as they were, with their namespace prefixes and any nested structure. Only the five properties on the form are replaced or removed.

    Why is the producer still named after I saved the copy?

    The producer records the software that generated the file and this tool leaves it alone, so the export still carries the original value. Only the five entries listed above change.

    Can I set a date on the copy?

    No. Creation and modification dates are read and reported, not written, because a malformed date entry makes some readers show an empty properties panel. Set a date in a desktop PDF editor if it matters.

    Why did the signed file lose its signature?

    A digital signature covers the exact bytes of the file. Saving an edited copy writes new bytes, so the old signature can no longer validate, and the report names the signature it found.

    Project manager: Tony Hines · Content updated 4 October 2026 · Report a problem