PDF Metadata Editor

It removes the obvious identifiers: the author name your word processor stamped in, the software that produced the file, the two timestamps, and the XMP stream that usually duplicates all of them. It does not touch what is on the pages. Names in the body text, EXIF baked into an embedded photo, tracked-change residue carried over from the source document and fonts that exist only on your machine all survive untouched. Metadata is not hidden content, deleting the author field does not redact a single word of the document. Treat this as one step of several.

No. The file is opened with pdf-lib's own metadata updating switched off, which matters more than it sounds: left on, pdf-lib stamps its Producer name and a fresh modification date into the document the moment it is parsed. That is why a file cleaned by some other tool so often ends up advertising the tool that cleaned it. Here the only things that change are the Info keys you edited and, if you leave the switch on, the catalog's XMP entry. Saving does invalidate an existing digital signature, but any edit to a signed PDF would.

An encrypted file is opened anyway, with a warning, because the Info dictionary is encrypted along with everything else: the fields may read as garbled characters and a saved copy can come out damaged. Take the password off with Remove PDF Password, which needs the password you open the file with, then edit the copy.

Nothing, an empty field deletes the key from the Info dictionary rather than storing a blank string.