Online tools that just work

I started Toolium because the existing free tool sites were maddening. You upload a PDF you want to merge. You wait for a progress bar. You sit through a 15-second ad. You get a watermarked output. You wonder, for the rest of the afternoon, whether the lease agreement you just uploaded is now sitting in some unknown server's cache forever. Every step of that flow is the wrong default.

The right default, in 2026, is that simple file operations stay on your own computer. Browsers can do an enormous amount of work locally now: pdf-lib handles PDF merging and splitting, browser-image-compression shrinks photos without round-tripping to a server, SheetJS reads and writes .xlsx, the Canvas API handles image format conversion, and Web Workers keep all of it from blocking the UI. The web platform got good enough years ago. Most tool sites just have not caught up.

So Toolium is 34 tools built on that principle. PDF merging, image compression, JSON formatting, password generation, regex testing, color picking, spreadsheet conversion, and on. Every one of them runs in your browser. None of them upload your file. None of them require an account. The trade-off is that you pay for them in CPU time on your own laptop, which honestly is the right trade.

The interesting work is not in the tools themselves. Those are mostly thin wrappers around well-built open-source libraries. The work is in the edge cases. Transparent PNG converted to JPG used to come out with a black background until I composited onto white before encoding. The Regex Tester used to freeze the tab on catastrophic-backtracking patterns until I moved execution to a Web Worker with a 1-second timeout. The Markdown Preview used to be vulnerable to XSS via embedded HTML until I added DOMPurify between the parser and the rendered output. Those fixes are what separate a tool you can trust from one you cannot, and they are the part of Toolium I am most proud of.

If you want to read more about how the tools work and what they do not do, the methodology page has the long version, including ten specific bugs I have shipped fixes for and the limits I will not try to push past. The About page has the project history. The blog has the educational content on the technical topics that come up while building this kind of thing.