What this does

Every page renders as a real thumbnail the moment a file lands in the drop zone, so you’re choosing based on what’s actually printed on the page rather than trying to remember whether page 7 was the cover letter or the invoice. Click a thumbnail and it dims with a red mark; or type a page range in the sidebar instead, which is usually faster for a long, scattered list like “2, 4, 9-14, 30”. Either input updates the other in real time, so you can switch between clicking and typing mid-task without losing your place. All of the rendering and rewriting happens with pdf-lib and pdf.js, running entirely in the browser tab you’re already in.

Two modes, two ways of thinking

Delete mode assumes most pages are staying and only a few are leaving — mark the outliers and everything else survives untouched. It’s the natural fit for cutting a handful of blank scans or a stray watermark page out of an otherwise clean document. Keep mode inverts that assumption entirely: mark only the pages you actually want, and everything else is discarded. That’s the better fit when you’ve been handed an 80-page appendix and only three pages of it matter — instead of clicking through 77 pages to remove them one at a time, you tag the three you need and let the rest go. Switching modes re-reads your existing marks rather than clearing them, so it’s easy to compare both readings before committing to a download.

Why this comes up more than you’d think

A signed page needs to move to the front before it’s emailed to a client. A confidentiality-marked page has to be stripped out before a document goes to a wider audience. A duplicate scan sneaks in when a stack of paper goes through a feeder twice. A ninety-page report needs to shrink down to the nine pages an executive will actually read. In every case the fix is the same shape — decide what stays, decide what goes, and produce a clean file without re-scanning, re-printing, or re-exporting the original from whatever software made it.

How the range field and clicks stay in sync

The page-range field understands the same shorthand most people already use in a printer dialog: single numbers separated by commas, and hyphens for a continuous run, so “2, 5-7, 12” marks pages two, five through seven, and twelve in one go. Typing a new range recalculates the marked set and redraws every thumbnail immediately. The reverse holds too — mark five thumbnails by hand and the field fills in with a tidy, collapsed range on its own, ready to copy out and reuse on a similar file later.

Batch & export

Multiple files can sit on the bench at once, each with its own independent set of marks and its own mode, so trimming a stack of scanned chapters doesn’t mean opening a fresh tab for every single one. A single file downloads directly as a ready-to-send PDF. Trim more than one at a time and the results are bundled into a single ZIP archive instead, so there’s one download prompt for the whole batch rather than a string of them.

Privacy & local processing

Nothing here touches a server. The file you drop is read directly into the browser’s own memory, rendered with pdf.js for the thumbnails you see, and rewritten with pdf-lib for the file you download — the bytes never leave the device they started on. That matters most for exactly the documents people tend to be most careful with: signed contracts, medical records, anything with a client’s name on it. Closing the tab clears the working memory completely, with nothing left behind to clean up.

Tips for best results

• Open the original once first if you’re unsure of the page order — thumbnails follow the file’s actual order, not any assumption based on the filename.
• For a long, scattered list of pages, typing a range is usually faster than clicking one by one; for three or four pages, clicking wins.
• Try Keep mode even when it feels backwards — for a short list of pages you want out of a long document, it’s often far less clicking than Delete mode.
• Use “Invert marks” to flip a Delete-mode selection into a Keep-mode one without retyping anything.

Frequently asked

Can I undo after downloading? The download is a new, separate file — your original stays untouched on disk, so re-uploading it is always the undo button.

Does the page order change? No. Pages that survive keep their original order relative to one another; nothing is reshuffled.

What happens to bookmarks or links? They carry over wherever the pages they point to still exist; a reference to a page that’s been removed is dropped along with it.

Notes & limitations

• You can’t mark every page in “Delete” mode, or none in “Keep” mode — a file needs at least one page left to be worth downloading.
• Password-protected PDFs can’t be previewed or processed here — you’ll see a clear error on that file’s card instead of a silent failure.
• Very large files (dozens of high-resolution scanned pages) will take a moment longer to render thumbnails; the file is still being read locally, not uploaded.