Editing here means adding your own content on top of the page: text boxes, freehand drawing, shapes, highlights and image stamps, placed wherever you need them and saved into a new PDF. It is the tool for filling in a form that has no form fields, annotating a drawing, initialling a document, marking up a proof or covering something with a box. What it deliberately does not do is rewrite the text that is already in the file. That distinction frustrates people until they know the reason, and the reason is structural: a PDF stores text as positioned glyphs with no concept of paragraphs or reflow, so editing a sentence in the middle of one would leave every following word exactly where it was. Everything happens in your browser — the document is never uploaded, whatever it contains.
Why existing text cannot simply be edited
A word processor stores a document as text with formatting rules and works out the layout when it displays. A PDF stores the finished layout: each glyph with a position, a font and a size, and no record of the paragraph it belonged to.
So changing a word means the words after it do not move, because nothing in the file knows they are part of the same sentence. Tools that advertise true text editing are reconstructing that structure by inference, which works well on simple, born-digital documents and poorly on anything else. Adding a text box over the top is the honest alternative, and for filling forms and correcting a line it is usually what you actually needed.
What you can add
Text boxes with a font, size and colour, for filling gaps and annotating. Freehand drawing for initials, ticks, circles and the sort of markup you would do with a pen. Rectangles and highlights for drawing attention or covering something visually. Image stamps for logos, a scanned signature or a photo.
Everything you add is drawn into the page when you save, so it prints and survives forwarding. It also stops being editable at that point — save a copy if the annotations are still in flux, and keep the original untouched so you can start again.
Covering is not removing
A filled rectangle over a paragraph hides it on screen and in print, and leaves the text underneath perfectly intact in the file. Copy and paste from the saved PDF and it comes straight back out, which is how confidential information gets published on a regular basis.
If the goal is to conceal rather than to annotate, use the redaction tool: it re-renders the affected pages so the covered content is genuinely absent from the file. The visual result looks the same; only one of them is actually true.