PDF tools, without uploading a thing

What you can do

1. Merge — drop two or more PDFs and combine them into a single file, in the order you choose.
2. Split — drop one PDF and download a .zip containing each page as its own PDF.
3. Rotate — drop one PDF and turn every page 90°, 180° or 270°.

Why do it locally

Most PDF tools upload your document to a server to process it. Contracts, statements, scanned IDs and medical records are exactly the files you should not hand to someone else’s server. Morphr edits the PDF right in your browser, losslessly, so text, fonts and images stay untouched, and nothing ever leaves your device.

Lossless by design

Merge, split and rotate work at the page level, so your pages are preserved exactly: no re-rendering, no quality loss, no re-compression. Rotation simply records the page orientation; splitting copies whole pages into new files.

Thirteen operations, none of which leave your device

This isn't one tool but a small suite, grouped by what it does to the file: rearranging pages (merge, split, rotate, extract or delete specific pages), protection (add a real AES-256 password, remove a PDF's print/copy/edit restrictions, or unlock one with its existing password), editing content (compress embedded images, flatten form fields, stamp a watermark, add page numbers), and converting between images and PDF pages in either direction (build a PDF from a set of images, or pull the images already embedded in a PDF back out). All thirteen run through the same wasm engine as the rest of Morphr — no operation here ever sends the PDF to a server.

Why merge, split and rotate don't touch a single pixel

These three work at the level of the PDF's own page objects rather than rendering pages to an image and reassembling them, which is what actually keeps them lossless: rotating a page just updates the orientation flag stored in that page's object rather than redrawing its contents at an angle, splitting copies whole page objects into new files, and merging concatenates page trees from multiple files into one. Text stays selectable, fonts stay embedded exactly as they were, and images inside those pages are copied as-is — nothing is decoded and re-encoded in the process.

"Remove restrictions" versus real password protection

These sound similar but do very different things. A PDF can carry permission flags (no printing, no copying text, no editing) that are enforced only by whichever reader application opens it; the file itself opens with no password at all, and "Remove restrictions" strips those flags so the file behaves as if they were never set. That's different from "Protect," which encrypts the PDF with a genuine AES-256 password: the file cannot be opened by any reader at all without the password. Protect additionally supports a separate owner password (controlling permissions) from the user password (controlling whether the file opens), matching how the PDF spec itself defines the two. If you leave the owner password blank, Morphr reuses the password you entered above it for both roles.

Compress, flatten and watermark

Compress only touches the raster images embedded in the PDF: it downscales and re-encodes them at the quality you choose on the slider, while text and vector content are left completely alone. That means a PDF built mostly from photos or scanned pages can shrink dramatically, while a text-only contract barely changes size at all, because there's rarely much to compress in it. Flatten is one-way: it bakes fillable form fields and any annotations into the page's permanent content, so the result looks identical but the fields can no longer be edited — useful right before sending a completed form, but not reversible afterward. Watermark stamps your text diagonally across every page (a "CONFIDENTIAL" or "DRAFT" mark, for example) directly into the page content.

Why this matters for the documents people actually run through it

PDF tools handle contracts, medical records, financial statements and scanned government IDs more than almost any other file type, exactly the kind of document you'd least want passed through a stranger's server just to rotate a page or knock a few megabytes off a file size. Doing it locally means none of that ever leaves the browser tab. One real limit worth knowing: "extract images" only pulls out images that are already embedded as separate image objects inside the PDF; it doesn't render or rasterize pages, and there's no OCR here. A scanned document saved as one full-page image per page will extract as those full-page images; a document with no embedded raster images at all (pure vector or text) has nothing for this operation to extract.