Three ways to change what’s on a PDF’s pages, in one editor. Annotate adds text notes, highlights and freehand drawings. Sign places a signature you draw or type, which you can move and resize. Redact removes content for good. Mix them in one pass — highlight a clause, sign the last page, black out an account number — and save once. Everything runs in your browser with pdf.js and pdf-lib; the document is never uploaded.
Annotate: a layer that sticks
Everything the annotation tools add sits on top of the existing page, which is the right model for filling a value onto a flat form, marking a correction, highlighting the clause that matters, or circling the wrong number. On save the marks are drawn into the page content rather than stored as comment objects, so they render in every viewer, print every time, and can’t be switched off in a comments panel. The page underneath — vector text, images, fonts — is kept, so nothing loses sharpness.
If the form has real, clickable fields, the PDF Form Filler writes into them properly instead of painting text on top. And if you need to change the words already in the PDF, that’s editing, not annotating: a PDF stores text as positioned glyphs, so the reliable route is converting to Word with the PDF Converter, editing there, and exporting back.
Sign: without the print-sign-scan ritual
The old way to sign a PDF was to print it, sign with a pen, scan it back and email the result. Here you draw your signature with a trackpad, mouse or finger — it’s cropped to the ink automatically — or type your name in one of four styles. Saved signatures wait in a palette: drag one onto the page (or tap it, then tap the page on a touch screen), move it into place, and resize it from its corner so it sits naturally on the line. Add the date next to it with the Text tool; it’s often required.
For everyday agreements a simple electronic signature like this is recognised in most jurisdictions, under laws such as the US ESIGN Act and the EU’s eIDAS regulation. It’s a visual signature, not a cryptographic one: documents that require a certificate-based (PAdES) or qualified signature need a dedicated signing service.
Redact: why “drawing a black box” became a scandal genre
Court filings, government releases and corporate disclosures have repeatedly leaked because someone drew an opaque rectangle over text in an editor — the text stayed in the file, one copy-paste away. Visual concealment and data removal are different operations, and only the second is redaction.
When you save, every page that has a redaction box is rendered, the boxes are painted onto the pixels, and the page is rebuilt from that image. The covered text, images and vector drawings no longer exist in the file. The trade-off is honest and limited to those pages: their remaining text can’t be selected or searched afterwards, and a document with redactions is reassembled page by page, so bookmarks and form fields don’t carry over. Pages without boxes are copied untouched.
Redaction failures are usually omissions — the third mention of a name on page 14. Before you start, list what must go and search for each item rather than scanning by eye:
- Direct identifiers — names, signatures, photos, employee or customer IDs
- Contact data — addresses, phone numbers, personal emails
- Government and financial numbers — national ID numbers, SSNs, IBANs, card numbers, tax IDs
- Case and account references that allow re-identification through another system
- Third parties who appear incidentally, and combinations like birth date + postcode + employer
Redaction is irreversible by design, so keep your unredacted original archived separately. For a photo or screenshot rather than a PDF, Redact Image does the same job.