PageBoard

Open a stack of PDFs, throw all their pages onto one shared canvas, and reorganize them like they were always a single document — drag pages between files, reorder, rotate, duplicate, delete, split, merge. Then save, and every file is exactly what you left it as. No uploads, no accounts, no watermarks: PageBoard runs entirely on your own machine.

It's free, open source, and built for the moment your PDF workflow stops being about one file and starts being about all of them at once — scanned chapters that need reordering, invoices that need splitting apart, forms that need to be assembled from three different sources. PageBoard turns that into drag & drop.

Why you'll like it

• One canvas for every open PDF. Documents sit side by side and pages move freely between them — drag a page out of one file and into another, drop it at the edge to spin off a brand-new document, select pages across file boundaries and act on all of them together.
• Two views, one workflow. Canvas mode keeps each document as its own column for spatial, per-file work; Grid view flattens everything into one tile wall when you just want to scan every page at once. Switch anytime without losing your place.
• Fast even when your PDFs aren't small. Pages only render once they're actually about to be visible, so opening a dozen large scans stays smooth instead of freezing the window.
• Nothing leaves your computer. No cloud processing, no upload step — PageBoard reads and writes the files already on your disk, and only the files you touch.
• Undo, always. Every page operation and every drag is undoable for the whole session, so exploring an idea never risks the original.
• Looks and behaves the same everywhere. Mac and Windows get pixel- identical rendering, since the whole UI runs on the same engine on both.

Screenshots

Canvas mode	Grid view

(Generated with placeholder content, not real documents.)

Documentation

User documentation (how to actually use the app) lives in docs/:

• Getting Started
• View Modes
• Organizing Pages
• Saving & Undo
• Keyboard Shortcuts

Feature set

• Open: multi-select file dialog, drag & drop from Finder/Explorer, detection of already-open files
• Two view modes: Canvas (per-document columns) and Grid (tile grid with adjustable column count), lossless switching between them
• Virtualized rendering: only visible pages are actually rasterized, even with many/large PDFs open at once
• Zoom & pan: anchored to the cursor position, plus focus mode (double-click a page)
• Page selection across documents: click, Shift+click (range), Cmd/Ctrl+click (toggle)
• Drag & drop of pages — within a document, between documents, or to the canvas edge (creates a new document)
• Reorder whole documents by dragging the section header, with a live preview
• Page operations: delete, 90° rotate, duplicate (context menu + keyboard shortcuts)
• Save overwrites the original file directly (no "save as"); empty documents can be restored or deleted on save
• Undo/redo for all page and move operations
• Close with a save/discard/cancel prompt for unsaved changes
• Robust error handling (corrupt/password-protected PDFs, missing permissions, …) instead of silent crashes
• Packaged .dmg/.exe builds via electron-builder, optional auto-update via GitHub Releases

Installing a downloaded build

The .dmg/.exe builds aren't code-signed yet, so macOS and Windows both flag them as coming from an unidentified source. That's expected, not a broken download — one extra step gets you past it:

• macOS: "'PageBoard.app' is damaged and can't be opened" is Gatekeeper's generic message for any unsigned, downloaded app, not an actual integrity problem. Fix in Terminal:

  xattr -cr /Applications/PageBoard.app

• Windows: SmartScreen shows "Windows protected your PC" — click More info, then Run anyway. Alternatively, right-click the installer → Properties → check Unblock → OK before running it, or run Unblock-File .\PageBoard.Setup.x.x.x.exe in PowerShell.

Build it yourself

There's no packaged download yet, but getting a running app out of this repo takes a couple of minutes. Requires Node.js 22 or newer (only for npm install/development tooling — the packaged app bundles its own Electron/Node runtime, so an end user never needs Node installed at all).

npm install
npm start           # launch the Electron app
npm test            # run the tests (node:test)
npm run screenshots # regenerate docs/screenshots/*.png
npm run build        # package for the current platform (→ dist/)
npm run build:mac    # .dmg
npm run build:win    # .exe (NSIS)

npm start opens a window in an empty starting state (drop zone + "Open…" button). PDFs can be opened via the button, via Cmd/Ctrl+O, or by dragging them in from Finder/Explorer. To debug the main and renderer processes together from VS Code, use the "Electron: Main + Renderer" launch configuration (see .vscode/launch.json).

Note for the VS Code integrated terminal: if npm start doesn't open a window, or electron --version prints a Node version instead of an Electron version, the culprit is the ELECTRON_RUN_AS_NODE environment variable inherited from VS Code. Fix: run without it — env -u ELECTRON_RUN_AS_NODE npm start.

Tech stack

• Electron – shared codebase for Mac and Windows (bundles its own Chromium, so rendering behaves identically on both platforms)
• pdf.js – rendering/display
• pdf-lib – page mutation (delete, rotate, duplicate) and writing. Note: the upstream project has had no release in some time; it's watched rather than treated as actively maintained, since it runs in the main process (full filesystem access) on data from files the app didn't create
• Drag & drop – native HTML5 Drag and Drop API, no library (dnd-kit is React-specific and doesn't fit this bundler-free setup)
• electron-builder + electron-updater – packaging for .dmg/.exe and optional auto-update via GitHub Releases, built by GitHub Actions on every release tag

Test PDFs

pdf-files/test-files/ contains sample PDFs for manual and automated testing, mirrored from py-pdf/sample-files — see NOTICE.md for licensing (CC-BY-SA-4.0, distinct from this project's own MIT license). The files are read-only so a save from the app can't accidentally overwrite them.

For testing the save flow (save/restore original/delete empty documents), use pdf-files/test-files-edit/: writable copies of a few sample files, kept in one central place instead of scattered ad-hoc copies. Deliberately not version-controlled (.gitignore), since the app modifies/deletes them during testing — just recopy from pdf-files/test-files/ if needed.

npm run screenshots regenerates the two screenshots above from the PDFs in pdf-files/screenshot-files/.

Contributing

Contributions are welcome — see CONTRIBUTING.md for setup, test strategy, and the project's scope.

License

MIT – open-source project, no monetization planned. Third-party licenses are listed in ACKNOWLEDGMENTS.md.