The free tool library that never asks for your files

Search for almost any everyday file task — "merge PDF," "compress image," "convert to Word" — and the results are dominated by sites that ask you to upload your file to a stranger's server first. Some queue you behind a daily limit. Some quietly attach a watermark unless you subscribe. Nearly all of them want an account, an email address, or both, before they'll do something as simple as combining two PDFs.

None of that friction is actually necessary. A modern browser is powerful enough to merge, split, compress, sign, and convert files entirely on its own — no server required. Open Tools Library is built around that fact. Every tool here does its work using JavaScript running on your own device, the same way a desktop application would, except there's nothing to install and nothing to sign up for.

The goal is simple: become the one link you bookmark for any everyday tool, instead of searching fresh every time and gambling on whichever ad-choked, upload-your-file site ranks highest that week.

When you drop a file into a tool on this site, your browser reads it directly into memory using the File API — the same browser-native mechanism any app uses to let you pick a file. From there, everything happens locally: pdf.js (the same rendering engine behind Firefox's built-in PDF viewer) decodes and displays PDF pages, pdf-lib writes new PDF bytes back out, and the browser's own Canvas API handles image decoding and re-encoding. At no point does your file get serialized into a network request.

Some tools go further. The PDF Editor and Sign PDF tools track every element you place in the exact coordinate space of the underlying PDF page, so what you see while editing is what ends up in the exported file — no separate upload-and-reprocess step to introduce drift between preview and output. The PDF to Word tool falls back to Tesseract.js, a WebAssembly build of the Tesseract OCR engine, when a scanned PDF has no text layer to extract — that's genuine optical character recognition, running as compiled WebAssembly in your tab, not a server-side call.

The only network request any tool makes is a one-time download of the OCR engine itself (from a public CDN, the first time you use OCR) — never your file. The full breakdown of exactly what each tool does and doesn't send anywhere is in our Privacy Policy.

Running and building a site costs something, even when every tool on it is free. We cover that with advertising — currently supplemented by Google Analytics to understand which tools actually get used, with Google AdSense planned to support it financially — never with subscriptions, never by selling data.

That second point isn't just a promise, it's a structural fact: since your files are processed entirely on your device and never reach our servers, there is no file data of yours for us to collect, analyze, or sell in the first place. The full detail on exactly what is and isn't collected — and how to opt out of analytics or personalized ads — is in our Cookie Policy and Privacy Policy.

This model doesn't change as more categories launch. Every tool that ships free — today, and in every category still marked "Soon" above — stays free.

Privacy on Open Tools Library isn't a toggle you switch on, or a setting buried in an account you had to create anyway — it's the architecture. There is no upload button that sends your file anywhere, so there's no "opt out of file sharing" checkbox to forget, and no server-side copy that could later leak, get subpoenaed, or be used to train a model.

That covers the actual files you process. For the handful of things the site does need to operate — anonymous traffic analytics, and eventually advertising — we've published our Privacy Policy, Cookie Policy, and Terms of Service in full, in plain language, rather than a wall of boilerplate designed to be skipped.

A tool that's private but wrong isn't actually useful — a merged PDF with a corrupted page, or a signature placed a few millimeters off, defeats the point just as much as a tool that uploads your file somewhere it shouldn't. So every tool here is built around exact coordinate math (PDF points converted precisely to on-screen pixels and back), not approximations, and tested by re-parsing its own output byte-by-byte to confirm the result actually matches what was requested — not just that it looked right on screen.

If you ever run into a tool that doesn't behave the way it should, that's genuinely useful to know about — tell us on the Contact page and we'll look into it.