Post History

[LibreOffice Draw](https://www.libreoffice.org/) (MPL-2.0) does that. It's hardly ideal, because unless you have all the standard Windows fonts (not actually tested!), almost all PDFs will have their text get rendered oddly including when you print. But you can perform the tasks of adding arbitrary text objects, images, etc. I use LibreOffice Draw when adding a scanned image of my wet signature and before scan-pdf.sh-ing it to make it look scanned so I don't have to kill a tree. Draw struggles with existing fillable fields, but that's hardly a problem when you can just add a text object on top of it!

[Atril](https://wiki.mate-desktop.org/mate-desktop/applications/atril/) (GPL-2.0) can fill out basic forms but has no keyboard navigation between forms and no additional arbitrary text/images.

[Firefox](https://www.mozilla.org/) (MPL-2.0) and its javascript tooling offers keyboard-based field navigation for the basic forms. No custom objects that I've found. It seems to handle the non-free font problem better somehow.