📄PDFZero - Free Open-Source PDF Editor.

Edit PDFs in your browser. No uploads. No task limits. No sign-up. Free.

---

Why PDFZero?

Many PDF tools charge monthly, cap file sizes, or limit what you can do on free plans. PDFZero keeps the core workflow simple: edit locally, keep your files on-device, and use the important tools without a paywall.

Demo

🌐 Try PDFZero Live

🔗 https://pdfzero-editor.vercel.app
Edit, organize, secure, and optimize PDFs directly in your browser - all FREE while keeping your files on your device.

---

---

Feature	Other PDF tools	PDFZero
Edit existing PDF text	Often paid or limited	Free
File size limits	Often capped	No file size limit
Daily task limits	Free plans may stop after a few tasks	No task limits
File privacy	Files may be uploaded to a server	100% local
Offline use	Usually browser or cloud-based	Works offline
Open source	Rare	MIT
OCR for scanned PDFs	Often paid	Free
e-Sign PDFs	Often paid	Free
Cost	Many plans charge monthly	Free

---

Features

Edit

• Edit existing PDF text - click any text block, edit in-place, and auto-detect the original font
• Add new text boxes anywhere on the page
• Change font family, size, color, bold, and italic
• Add, replace, and remove images

Organize

• Merge multiple PDFs with drag-to-reorder
• Split PDF by page range
• Reorder pages via drag-and-drop
• Rotate individual pages
• Extract specific pages

Optimize

• Compress PDF with browser-native object stream compression
• PDF/A compliance check

Secure

• Password protect with AES-256
• Remove existing passwords
• Redact sensitive content permanently
• Add text watermarks

Smart

• OCR for scanned and image PDFs with Tesseract.js, running offline
• AI font matching to keep text edits visually consistent

---

Tech Stack

Library	Purpose
pdf-lib	PDF creation, modification, export
PDF.js	PDF rendering and text extraction
Tesseract.js	OCR for scanned PDFs
React	UI framework
Zustand	State management
Vite	Build tool

Zero backend. Zero tracking. Zero analytics. 100% browser-native.

---

Getting Started

git clone https://github.com/bevinkatti/pdfzero.git
cd pdfzero
npm install
npm run dev
npm run build

---

Architecture

src/
  components/
    editor/          # PdfCanvas, TextBlock, AnnotationLayer, Toolbars
    layout/          # Navbar
    ui/              # DropZone, shared components
  lib/
    pdfRenderer.js   # PDF.js wrapper - render pages, extract text
    pdfExporter.js   # pdf-lib wrapper - export, merge, split, etc.
  pages/
    Landing.jsx      # Marketing landing page
    Editor.jsx       # Main PDF editor
    Tools.jsx        # Individual tool UIs
  store/
    pdfStore.js      # Zustand global state
  styles/
    globals.css      # Design system tokens

Text editing architecture

PDFZero uses a layered editing model:

PDF bytes
  -> PDF.js render + text extraction
  -> canonical text run metadata
  -> browser overlay editor
  -> layout-fit/export planner
  -> pdf-lib browser export
  -> future PDFium/MuPDF advanced engine

Current browser path:

1. Render - PDF.js renders each page to a high-resolution canvas.
2. Extract - PDF.js extracts text runs, transforms, font ids, approximate font names, color, baseline, ascent/descent, and edit boxes.
3. Model - PDFZero stores original run metadata: text, bbox, baseline, font fallback, color, line height, estimated glyph advances, and max edit dimensions.
4. Overlay - Editable DOM text is positioned over the PDF raster and uses local background sampling to hide the original text while editing.
5. Fit - On export, edited text is laid out line-by-line and fit back to the original run width using conservative character spacing and small size adjustment.
6. Export - pdf-lib writes background patches, replacement text, annotations, page operations, and document tools.

This is not yet full Acrobat/Foxit-style object editing. The current path is an overlay-and-repair browser exporter. The production-grade target is an advanced engine that can inspect and rewrite PDF page objects directly:

src/pdf/
  engines/
    pdfjsEngine.ts       # Browser render/extract fallback
    pdfiumEngine.ts      # Planned object-level editor
    mupdfEngine.ts       # Optional native/WASM alternative
  model/
    TextRun.ts           # glyphs, fonts, colors, matrices, resources
    FontResource.ts
    EditOperation.ts
  layout/
    glyphMetrics.ts
    fitText.ts
    paragraphReflow.ts
  background/
    renderWithoutText.ts
    inpaintPatch.ts
  export/
    exportPlanner.ts
    pdfLibOverlayExporter.ts
    objectRewriteExporter.ts
  repair/
    visualDiff.ts
    autoFitRepair.ts

The browser-only implementation can be excellent for simple and moderately complex PDFs. True high-level editing for embedded subset fonts, object removal, kerning-preserving replacement, complex scripts, and image/gradient background reconstruction requires PDFium, MuPDF, or another real PDF content engine.

---

Roadmap

• PDF rendering and text extraction overlay
• Add new text boxes
• Drag-and-drop text positioning
• Annotations (highlight, redact, shapes)
• Merge, split, compress tools
• Watermark, rotate, page management
• Preserve richer original text-run metadata for export fitting
• Fit edited text back to the original run width during pdf-lib export
• v1.1 - OCR via Tesseract.js with searchable text layer
• v1.1 - Visual export diff for edited regions
• v1.1 - Packaged fallback font registry with width-vector matching
• v1.1 - Image add/replace/remove
• v1.1 - e-Sign with canvas signature pad
• v1.2 - Paragraph grouping and multi-line reflow
• v1.2 - PDF to Word/DOCX export
• v1.2 - Form filling and flattening
• v1.2 - Batch processing
• Advanced - PDFium/MuPDF object-level text replacement
• Advanced - Embedded font reuse and kerning-preserving export
• Advanced - Background reconstruction by rendering pages with target text objects removed

---

Contributing

PRs are very welcome. Please open an issue first for major changes.

npm install
npm run dev

---

License

MIT Copyright PDFZero contributors

---

If you find PDFZero useful, consider giving a ⭐.