About PDF Editor
Annotate any PDF directly in your browser, page by page. The PDF Editor renders each page to a canvas using pdfjs-dist, lets you draw arrows, rectangles, circles, highlights and redaction bars on top, add text labels at any position, and undo / redo per-page. On export, your annotations are baked into the PDF as additional content streams via pdf-lib — the original text, images, and existing PDF annotations are all preserved untouched, and the overlay just sits on top. The whole pipeline runs locally; your PDF never leaves the device. The honest scope: this is an additive editor — for changing existing text content (replacing words, editing paragraphs in place) the PDF format is too lossy to do reliably in a browser; PDF→Word is the right round-trip there.
- No uploads
- Browser-only
- Works offline
- 100% free
How it works
- 1
Drop your PDF
Up to 100 MB. The first page renders immediately. Use the page navigation to jump anywhere in multi-page documents.
- 2
Pick a tool, click and drag
Arrow, rectangle, circle, highlight, text, redact (solid bar). Each annotation is per-page — switching pages shows a clean canvas with that page's annotations.
- 3
Export the edited PDF
All your annotations bake into the PDF. The output preserves every byte of the original content streams; your overlay is a new content stream stacked on top.
What "editing" a PDF really means
A PDF is a layout format, not a document model. When a word processor exports one, it doesn't store sentences and paragraphs — it stores instructions like "draw this glyph at this x/y position in Helvetica 11pt." There is no reflowing text underneath; there are positioned shapes that happen to look like words. That is why true in-place text editing of an arbitrary PDF is genuinely hard, and why most honest PDF editors — this one included — edit by adding a new layer on top of the page rather than rewriting the original content stream.
The practical upshot: SnapToolz lets you stamp new text, draw rectangles and lines, highlight regions, and place freehand ink anywhere on a page. You are annotating — adding to the page — not splicing into the existing text run. For most real tasks (filling a form field that isn't interactive, crossing out a clause, adding a note, blacking out a number) that is exactly what you want, and it's faster and safer than a round-trip through an editor that re-renders the whole document.
What you can add, and how it behaves
- Text boxes — choose a font, size and colour; the text becomes part of the page on flatten. Match the original font visually, but don't expect kerning to line up perfectly with existing text — you're overlaying, not inserting.
- Highlights — a translucent colour band over a region. Unlike a true text highlight in a reader, this is a drawn rectangle, so it covers whatever sits beneath it regardless of word boundaries.
- Shapes and lines — rectangles, lines and arrows for redaction boxes, signature underscores, or call-outs. A filled black rectangle is a visual cover, not a redaction (see below).
- Freehand ink — draw with a mouse, stylus or finger for a signature or a circle around a figure. For a dedicated signing flow with a saved signature, use Sign PDF.
Flattening: when your edits stop being editable
Annotations live in a separate layer that a reader can, in principle, select, move or delete. Flattening merges that layer into the page so the marks become permanent, non-selectable page content — exactly like ink on printed paper. SnapToolz flattens on export, which is what you want for anything you're sending out: the recipient sees your edits baked in and can't peel your highlight off to read what's underneath.
There is a catch worth knowing. If you placed a black box over sensitive text expecting it to redact, flattening does not delete the text beneath — it only hides it visually. The original characters remain in the content stream and can be recovered with copy-paste or a text extractor. Genuine redaction means removing the underlying text, not covering it. If you're cleaning a document before sharing, follow the full process in how to privacy-clean a file before sharing.
Common mistakes when editing PDFs
- Treating a black rectangle as redaction — the hidden text is still selectable. Remove the text, don't just cover it.
- Trying to retype an existing paragraph — you can't edit the original text run in place; you'd cover the old text with a new box, which rarely aligns. For heavy text changes, edit the source document and re-export.
- Forgetting an encrypted input — a password-protected PDF can't be annotated until it's opened. Clear a known password first with Unlock PDF.
- Leaving identifying metadata on the file — your edits are visible, but the document's author, software and timestamps aren't. Overwrite them with Edit PDF Metadata before you send.
Frequently asked questions about PDF Editor
How does this differ from the existing /pdf/sign, /pdf/watermark, /pdf/page-numbers tools?
Those tools are templated — sign once, stamp the same watermark on every page, add page numbers via a formula. The PDF Editor is for free-form per-page annotation: this paragraph gets an arrow, that one gets a redaction bar, this page gets a note. They're complementary.
Can I edit existing text in the PDF?
No. Replacing existing text in a PDF requires reconstructing the original font and layout, which is unreliable in any browser-side tool. For text edits, use PDF→Word to round-trip into an editable .docx, edit there, then Word→PDF to export.
Does the editor work for scanned PDFs?
Yes — annotations render on top of the page image. The underlying text isn't editable (because it's an image), but you can mark it up exactly the same way.
Why no blur tool here, when the Screenshot Annotator has one?
Blurring requires re-encoding the page region as an image, which would destroy text editability and bloat the file. For privacy-sensitive areas use the Redact tool — it's a solid bar that's irreversibly opaque, which is actually more secure than a blur (deblur attacks exist, opaque rectangles are unrecoverable).
What about rotated pages?
The editor warns when the source PDF has rotated pages. Annotations may be mispositioned because the editor's coordinate system doesn't currently rotate with the page. Run the PDF through /pdf/rotate first to normalise rotation, then edit.
Are annotations saved between sessions?
No. Annotations live in your browser's memory while the page is open and are baked into the exported PDF. Close the tab without exporting and they're gone. This is intentional — no auto-save means no data persisted on your device beyond the PDF you explicitly download.
Privacy, offline use, browser support, and pricing questions are answered on the site-wide FAQ.