Single shows one page at a time, centered.

Continuous scrolls every page in one vertical strip.

Two-Page shows facing pages as spreads, like an open book.

Grid lays out every page as a thumbnail wall for a whole-document overview.

Switch from the toolbar or with F5 to F8. Every mode shares the same annotations and zoom, so nothing shifts when you change layout.

Open as many PDFs as you like, each in its own tab.

Ctrl+Tab and Ctrl+Shift+Tab cycle through them.

Ctrl+W closes the current tab; Ctrl+Q closes them all.

Each tab remembers its own zoom, view mode, page, scroll position and tool, so switching back drops you exactly where you left off.

The active tab always stays visible; when the strip fills up the rest move into an overflow menu.

Press F10 to show two documents side by side in one window - each pane is its own card, and the focused one carries an accent ring so it is obvious which pane the toolbar, sidebar and page list are acting on.

Click either pane to move focus, or drag its resize handle to size it; neither pane can be squeezed below a readable width.

Drag a tab from one pane's strip onto the other to move it across. F10 again closes the split.

A maximized or snapped window splits its space evenly instead of squeezing the second pane down to its minimum.

A dockable sidebar with two tabs: Pages (live thumbnails) and Outline (the PDF's bookmarks). Toggle the whole thing with Ctrl+B.

Reorder pages by dragging a thumbnail; a drop indicator shows where it will land.

Multi-select pages, then rotate, extract or delete them all at once.

Right-click a thumbnail for the full page menu: Insert blank page, Duplicate, Rotate clockwise / counter-clockwise, Move up / down, Extract pages to a new PDF, Delete pages, or Stamp pages.

Drag the slim splitter to resize the sidebar, and dock it on the left or right from the sidebar's right-click menu, or press Ctrl+Shift+B - it remembers your choice. A toggle strip collapses it to a thin rail.

The Outline tab shows the PDF's bookmark tree, nested and expanded; click any entry to jump to its page. It sizes itself to the widest bookmark the first time you open it.

Ctrl+F opens a small floating search bar you can drag anywhere.

It searches the whole document as you type (after two characters), case-insensitively, and can match phrases that run across adjacent words.

Enter jumps to the next match, Shift+Enter to the previous, wrapping around the document; the status reads current / total.

Every match stays highlighted while you scroll - the current one brighter and thicker, the rest dimmer - and the highlights survive zooming and re-rendering.

Zoom runs from 5% to 500% in 15% steps; Ctrl+0 snaps back to 100%.

Ctrl + mouse wheel zooms around the point under your cursor, so the spot you are reading stays put.

The zoom box also offers Fit Width and Fit Page, computed exactly to the page size.

In Grid view the wheel changes the number of columns instead, so you zoom by fitting more or fewer pages across the window.

Clickable links already in the PDF light up with a hand cursor.

Left-click follows them - an internal link jumps to its page, an external one opens in your browser.

Right-click a link to copy its URL or remove it.

Press T (or toolbar slot 1), click anywhere, and start typing a note. Enter saves it, Escape cancels.

While it is selected, the floating text bar sets the font, point size, color, an optional background fill, and Bold / Italic / Strikethrough / Underline.

Drag the box's corner handles to set its width - the text wraps to fit.

Double-click a note any time to re-edit the wording.

This is the part most editors fake. Double-click a line of real PDF text with the Select tool, and KillerPDF reads the actual words underneath and drops two linked annotations:

1. A Cover - an opaque box that blocks out the original words. It does not just slap down white: it samples the page just above and below the line and matches that background color, so a cover over a colored header blends in.

2. A Text box on top, pre-filled with the detected words and starting in the same color and size - the font and point size are read straight from the original letters.

The two are locked together as a pair. Selecting either half dash-outlines its partner, so you never lose track of the other piece.

Make sure the old words are really gone. Click the cover (the background area) and use its settings bar to change its color, or drag its corner handles so it fully covers the original. A cover is always 100% opaque on purpose - it can never go see-through and let the old text ghost back.

Then resize and restyle the new text like any text box: drag the handles for width, change the font, size, color or weight.

Right-click either piece to Select Cover / Select Text - this jumps the selection to the other half - or to Unpair them into two independent annotations.

The whole edit counts as a single undo, and on save it flattens to a clean filled box with your new text on top.

On a scanned page with no text layer, you get an empty cover and text box at the click point: type your replacement, then drag the cover over the scan to white the old text out by hand.

See how the cover-pair is built, with a diagram, on the Technical page →

Press H and drag to lay a highlight over text. Color and opacity live on the bar; the default is yellow at low opacity.

The Highlight tool has an Eraser toggle, and it works two ways.

Brush: drag over a highlight and it carves a smooth, round-edged gap out of it - one anti-aliased shape, not a blocky strip or seam. A circular brush footprint shows where the next pass will land.

Box: drag a rectangle (shown as red marching-ants) and it deletes whole annotations inside the box.

The brush only carves fill highlights and splits ink strokes; text, images and signatures are left untouched.

Press D for freehand Draw - round-capped ink that keeps drawing stroke after stroke without re-arming.

Draw has its own Eraser toggle: a brush (with a translucent grey footprint preview) that passes over your ink and splits the strokes wherever it touches, rubbing out just those parts rather than deleting whole strokes.

Press L (or U) for the Line tool, a straight two-point segment.

Turn on Level and the line snaps to perfectly horizontal or vertical, whichever your drag is closer to.

All three carry color, opacity and width on their bar.

The Shapes tool draws rectangles, ellipses and free-form polygons, each with an optional fill - pick the sub-mode and toggle Fill on the bar.

Box keeps the classic drag-a-rectangle gesture. Ellipse drags the same way as a closed outline.

Freeform places points click by click: click the first point again (its target lights up when you are close) or double-click to close the polygon, Esc cancels, Backspace removes the last point.

Shapes share the draw bar's color, size and opacity, and move, resize, flatten and print like any other drawing.

Press I, click the page, and pick an image file.

It lands at half size; drag a corner handle to resize it.

You can also right-click the page and choose Image to drop one exactly where you clicked.

With the Select tool, drag an empty area to rubber-band select everything the box touches. Because the box floats above the pages, one drag can grab annotations across several pages in Continuous or Grid view.

Shift-click adds or removes individual annotations from the selection; Ctrl+A selects them all.

Right-click a multi-selection and choose Group to bind them together. After that, clicking any one moves, copies or deletes the whole group as a unit.

Ungroup dissolves the group; Remove from group drops just one member.

Use Raise and Lower on the right-click menu to change which annotation sits on top where they overlap.

A selected annotation gets an accent outline and four corner handles. A text box's handles set its width; images, signatures and highlights resize freely. The outline and handles stay the same on-screen size no matter how far you zoom.

Press C to crop. A box appears inset from the edges with draggable corner handles, plus exact X / Y / W / H fields.

Switch the units between points, inches or %. Pressing Enter in a field both sets and applies the value.

Crop the current page, a typed range like 1-3,5, or tick All for every page.

It writes a real /CropBox and /TrimBox into the PDF and is fully undoable; Remove crop puts it back.

Quick 90-degree turns live on the right-click menu and the R key.

The Transform window does more: rotate by any angle, scale, flip horizontally or vertically, and straighten a crooked scan by drawing a level line across it.

Correct trapezoidal distortion: expand Perspective, turn the correction on, then drag the four green corner handles to the corners of the rectangular object in the page. This is useful for photographed documents, signs and screens that appear narrower on one side.

The four sides must form a valid shape and must not cross. Use Reset in the Perspective section to return the handles to the page corners, or Reset all to clear every transform setting.

The preview updates as you move the handles. Choose Apply to rebuild the page at full resolution. The edit is physical, includes the page annotations, refreshes every visible view immediately, and can be reversed with Ctrl+Z.

Insert or append blank pages (A4), delete, duplicate, move up or down, and extract or split pages out to a new PDF.

Most of these live on the thumbnail right-click menu; drag thumbnails to reorder.

Structural edits are undoable, and your annotations follow their pages.

Open images - JPG, PNG, BMP, GIF, TIFF - as PDF pages; multi-page TIFFs and GIFs come in frame by frame.

Drop a folder and it recurses through everything inside; drop a .zip and it extracts first.

When you drop several files at once, KillerPDF asks whether to merge them into one PDF or open each separately (up to 50 at a time).

Press S to open one window with a live preview that steps through your pages, holding two independent jobs.

Page numbers: set a start number and a format using {n} for the current page and {N} for the total - for example Page {n} of {N}.

Choose the font size, color, a page range, and one of eight positions (or drag a custom spot).

Turn on Mirror and the numbers flip to the outer edge on facing pages, like a printed book.

Watermark: stamp text or an image across any page range, with adjustable size, angle (-90 to 90), opacity and position.

Stamps draw beneath your annotations, count as a single undo, and you can double-click one to edit it again.

Interactive form fields - text, multi-line, dropdowns, checkboxes, radio groups and signature fields - become editable right on the page, in every view mode.

While a text field is focused, a small font-size stepper ( - N + ) lets you size the text, and it is saved into the field.

Fields show a faint cream fill with an accent border when focused.

On save your entries are written back into the PDF and the fields stay real, editable form fields - they are not destructively flattened.

Sign a document cryptographically with a certificate from a .pfx / .p12 file or the Windows certificate store.

It uses SHA-256 and embeds the whole certificate chain.

Add an optional reason, location and contact, and it writes a signed copy alongside the original.

Press G to open the signature popup. Draw a signature on the pad with three pen widths (thin, medium, thick), or import an image of one.

It keeps two kinds: full Signatures and shorter Initials, saved for reuse so you only draw them once.

Drop a saved signature anywhere on the page, or straight into a form sign-field - it centres and scales to fit, and remembers your pick for that kind of field.

A signed field then offers Change or Remove.

KillerPDF reads text off scans and images with on-device OCR - nothing is ever uploaded.

OCR Page (Ctrl+Shift+O) copies the current page's text to the clipboard and reports a confidence figure.

OCR Region (Ctrl+Shift+I) lets you drag a box to OCR just that area to the clipboard.

Make Searchable PDF adds an invisible real-text layer aligned over the words, so search and text-select start working on a scan.

Extract All Text dumps the whole document to a .txt or .md file.

Languages & models: ten languages - English, Spanish, French, German, Czech, Japanese, Turkish, Bengali, and Chinese (Simplified and Traditional), the same ten the interface speaks - download on demand the first time you pick them from the OCR language menu.

Each language offers a Standard model (smaller, faster) or a High Quality model (larger, more accurate), and once downloaded they are cached so you only fetch them once.

Pages are rendered at about 300 DPI for accuracy.

Ctrl+P opens a real print preview.

Print 1, 2, 4, 6 or 9 pages per sheet (N-up), with scale (Fit / Actual / Custom %), a 9-way alignment, margins, a page range, color or black-and-white, copies, and two-sided output when your printer supports it.

A subset selector under Pages prints all pages, odd only, or even only - applied on top of the page range, with the preview following along. Print the odds, flip the stack, print the evens: manual duplex for printers without a duplexer.

Output is rendered at a true 300 DPI.

If the document has annotations or stamps, they are flattened onto a throwaway copy for printing, so your live document is untouched. You can also save a flattened, uneditable copy.

F4 opens Document Info.

View and edit the PDF's title, author, subject, keywords and creator, all saved back into the file.

KillerPDF sends no telemetry and phones home for nothing except the update check against GitHub. Everything - OCR included - runs on your machine.

On a shared machine, the recent-files list is the one trace of what you opened. Clear list lives on the start screen's Recent panel and in the Open dropdown, and the About window (F12) has a "Don't remember recently opened files" toggle next to Clear all Data - turning it on also empties the existing list.

Clear all Data in the About window wipes every stored setting and saved signature in one go.

Six themes - Dark, Light, High-Contrast, Blood, Greed and Cyanotic - switch live, and the three base themes each take one of six accent colors.

The sidebar docks on the left or right.

The interface is translated into ten languages.

Invert document colors: the moon toggle at the bottom of the sidebar rail (or the N key) renders the document with inverted colors for dark-mode reading. Display only - saving, printing, exporting, OCR and the thumbnails all keep the document's true colors - and remembered across launches.

App size: scroll the mouse wheel over the title-bar logo (or Ctrl+Shift and + / -, Ctrl+Shift+0 to reset) to scale the whole interface from 70% to 250% - toolbar, sidebar and tabs, while the document keeps its own zoom. Made for tired eyes and high-DPI laptops alike.

Every core operation also runs headless from PowerShell or cmd - no window opens, and it works even while KillerPDF itself is running.

Run the portable EXE directly, or the installed copy at %LOCALAPPDATA%\Programs\KillerPDF\KillerPDF.exe.

Exit codes are the scripting contract: 0 success, 1 the operation failed, 2 bad usage. Check $LASTEXITCODE in PowerShell.

KillerPDF.exe --help prints the full reference; --version prints the version. Because KillerPDF is a windowed app, console text can interleave with your prompt - that is normal.

Page ranges are written like 1-3,5,9-12. Encrypted inputs are unlocked automatically where possible; pass --password <p> for password-protected files.

KillerPDF.exe --merge out.pdf a.pdf b.pdf scan.jpg merges PDFs and images into one file, in the order given.

Merges rewrite named-destination links, so internal cross-references and TOC links keep working in the combined file.

KillerPDF.exe --extract-pages in.pdf 1-3,5 out.pdf pulls a page range into a new PDF.

KillerPDF.exe --split in.pdf pages\ writes one PDF per page, named in-page-001.pdf and so on.

KillerPDF.exe --decrypt locked.pdf open.pdf removes encryption losslessly (bookmarks and forms survive). Add --password <p> if the file needs a password to open.

KillerPDF.exe --flatten in.pdf flat.pdf rasterizes every page into an uneditable PDF, the same 150 DPI output as Save Flattened in the app. --dpi overrides the resolution.

KillerPDF.exe --to-image in.pdf imgs\ --dpi 300 --format jpg renders pages to PNG (default) or JPG files. --pages limits the range.

Rotated pages come out in their viewed orientation in all three.

KillerPDF.exe --print in.pdf --printer "HP LaserJet" --pages 1-4 --copies 2 prints silently at 300 DPI, fit-scaled and centered. Without --printer it uses the Windows default; a partial printer name matches too, and a bad name lists what is available.

KillerPDF.exe --ocr scan.pdf searchable.pdf --lang eng lays an invisible, selectable text layer over a scan - the same on-device OCR as the app, nothing uploaded.

Language models (eng, spa, fra, deu, ces, jpn, tur, ben, chi_sim, chi_tra) download automatically on first use and are cached.

KillerPDF.exe --batch-resave inDir\ outDir\ --log report.csv resaves a file or a whole folder tree through the standard open/save pipeline, mirroring the folder structure.

Each file reports OK, SKIP (with a reason), or FAIL, to the console and the optional CSV log.

This powers KillerPDF's standards-conformance testing: resave a validation corpus, run veraPDF over input and output, and diff the results with validation/Compare-VeraPDF.ps1 from the repo. The claim it verifies: a KillerPDF save does not degrade PDF/A conformance.