AI field scan
Upload a form and the scan finds text boxes, checkboxes, dates, amounts, and signature lines automatically — importing values from forms that are already filled in, even locked ones.
Loading editor…
A complete PDF form editor that never uploads your document
The Novus PDF editor combines an AI form-field scanner with a full set of manual tools — text, numbers, checkmarks, crossmarks, drawn signatures, text-aware highlighting, and white-out — plus all ten built-in document tools for merging, splitting, organizing, rotating, numbering, protecting, and unlocking. Everything runs locally in your browser, which makes it a safe place for the documents that actually matter: government forms, medical paperwork, contracts, and financial records.
What the editor does
Fill any field
Type into detected or manually placed text and number fields with export-safe fonts (Helvetica, Times, Courier), adjustable sizes, and precise numeric positioning.
Checks and crosses
Toggle checkboxes with one click, or place crossmarks where a form expects an X. Autofill mode jumps to the next empty field so you can Tab through a document.
Real drawn signatures
Sign with a mouse, finger, or stylus. New signatures export as normalized vector strokes and stay sharp when resized; legacy transparent bitmap signatures remain supported.
Text-aware highlighting
Hover to preview, then click a line or drag across a passage — the highlighter snaps to the words like a real marker, with an area mode for regions and images.
Erase and white-out
Cover unwanted text or areas with a page-colored box that hugs exactly what you select. Match the cover color to the paper with one click.
Document tools built in
Merge, split, organize, rotate, number, compress, flatten, protect, unlock, run OCR, export PDF pages as images, or build a PDF from PNG and JPEG files without sending the document to a server.
Private by architecture
The PDF is parsed, edited, and rebuilt inside your browser tab. Password-protected files are decrypted in memory with your password — nothing is ever uploaded.
Will it work on my PDF?
Every PDF is a container, and what is inside decides how much of the work the editor can do for you. The five cases below cover essentially everything people open here. In all of them the manual tools work identically — the difference is only how many fields you get for free.
- Digital forms with real form fields
- Tax packages, bank applications, HR onboarding, and most government forms distributed as PDFs. These already carry interactive widgets, so the scan maps them one-to-one: every box becomes an editable field in the right place and the right size, and any values already filled in are imported so you can correct them instead of starting over. Best case — usually fillable end to end with no manual work.
- Static digital forms (lines and boxes, no fields)
- A document exported from Word or InDesign that looks like a form but has no interactive layer. The page still has real, selectable text, so labels like Name, Date, and Signature are detectable and fields get placed near them. Expect a good starting layout with some positions to nudge, especially in dense multi-column tables. Usually works — plan on adjusting a few placements.
- Scanned or photographed pages
- A phone photo, a flatbed scan, or a fax starts as a picture of paper with no text under the ink. Every manual editor tool still works. The separate local OCR tool can add a searchable text overlay in English, French, or Spanish, with low-confidence results held for review; it does not automatically turn recognized words into form fields. Local OCR is available; form fields still need manual placement.
- Password-protected documents
- Enter the document's password when prompted. It is used in your browser's memory to decrypt the file for viewing and export, and is never stored or transmitted. Permission-restricted files that still open for viewing behave the same way. Supported, with your password.
- XFA and flattened documents
- XFA forms (an Adobe-specific dynamic format used by a few older enterprise and government systems) render as their static fallback page, and flattened PDFs have had their fields permanently baked into the page. In both cases the interactive layer is gone or unreadable, so the page behaves like a static form. Opens and edits as a static page.
Deeper walkthroughs: how the AI field scan works, making a static form fillable, filling and signing a scanned PDF, and a full government-form run-through.
What the editor will not do
Worth knowing before you rely on it for something that matters.
Cover is recoverable; Remove rasterizes
Cover paints a page-colored rectangle over the area you select, so the PDF content underneath remains recoverable. Remove rasterizes each affected page so covered content is absent from the export, but selectable text, links, form fields, annotations, and tags on that page are lost. Neither mode is a certified redaction workflow; read the walkthrough below and inspect the export before sharing it.
OCR does not create form structure
Field detection reads the document's own structure and text, while the local OCR tool adds a searchable overlay to an image-only page. OCR supports English, French, and Spanish and requires review; it does not automatically create form fields or a tagged-PDF structure tree.
Very large documents are limited by your device
There is no server doing the work, so a 400-page file is bounded by your browser's memory rather than by a plan tier. Big documents open, but rendering and export take longer on a phone than on a laptop. Splitting a long file first and merging afterwards is usually faster.
Signatures are drawn, not certified
A newly drawn signature is embedded as normalized vector strokes, while legacy transparent bitmap signatures remain supported. Neither is a cryptographic digital signature backed by a certificate authority, so neither satisfies a workflow that specifically requires PKI-based signing.
On that first point in particular, read how to redact sensitive information in a PDF — it explains why a black box is not a redaction and what to do instead. For marking up rather than hiding, the highlight and white-out guide covers where each tool is the right choice.
What comes out at the end
Export writes a new PDF rather than a picture of one. Text you typed is embedded as real text in a PDF-safe font (Helvetica, Times, or Courier), so it stays selectable, searchable, and crisp at any zoom — the same as the document’s original text. The underlying pages are copied through untouched, which means images keep their resolution and vector artwork stays vector. Highlights export with the same multiply blending you see on screen, so the text under them remains readable rather than washed out, and a drawn signature keeps the exact proportions you drew it at wherever you moved or resized it.
Two practical notes. First, the exported file is unencrypted even if the original was password-protected — that is deliberate, so you always get a file you can open; run it through Protect PDF afterwards if the recipient needs it locked. Second, printing goes through the same export path as downloading, so the printed page matches the download exactly. Before submitting anything official, open the finished file once in a normal PDF viewer — it takes ten seconds and catches the occasional field you meant to fill and did not.
More on the finishing steps: signing a PDF electronically and building a submission-ready packet.
How it works
- 1Upload any PDF — the AI scan runs automatically and places editable fields over everything it detects.
- 2Click a field to fill it; press Enter to jump to the next empty one. Switch to manual tools for anything the scan missed.
- 3Draw your signature, highlight or white-out text, and fine-tune positions with the numeric controls.
- 4Export a finished PDF or print directly. Exports are vector-based: original text stays selectable and crisp.
Want the full walkthrough? See how it works, the guides, or the mini tutorials.
PDF Editor questions
Is the PDF editor free?
Yes — completely. There are no accounts, watermarks, or export limits. The site is supported by advertising.
Can it open password-protected PDFs?
Yes. Enter the document password when prompted; it is used only in your browser's memory to open and export the file, and is never transmitted.
Does it work on scanned documents?
Scanned (image-only) PDFs open and display normally, and every manual tool works on them. The AI scan needs digital text to detect fields, so on pure scans you place fields by hand.
Will the exported file match what I see?
Yes. Exports reproduce the on-screen result: highlights use the same marker-style blending, signatures keep their drawn proportions, and text uses the same PDF-safe fonts.
Need a single-purpose transform instead? The PDF power tools handle merging, splitting, organizing, rotating, numbering, protecting, and unlocking in focused workspaces.