We build small, free browser tools for the boring parts of creative and technical work, and we write about the workflows they came from. No accounts, no installs, no trial that expires.
Every tool here started as a problem in our own week rather than a gap in a market. The Premiere Pro Scroll Amplifier exists because a scroll wheel that jumps the timeline to maximum zoom ruins an edit, and the fix turned out to be a mouse setting almost nobody knows about. TailWinks exists because Tailwind v4 moved its palette to OKLCH while Figma variables store sRGB, so copying one into the other quietly stopped working.
The extractors came from the same place. Pulling the text out of a page, a PDF or a deck so you can actually read or compare it is a job that happens constantly, takes ten minutes at a time, and nobody counts. Each tool does one step of that and stops.
View Our Tools
Ten tools, all free. Seven run in the browser, and the other three install into the app they work with. They fall into two groups.
None of them require an account and none of them have a paid tier waiting behind a limit. If a tool cannot do something, the page says so rather than making you find out.
How We WorkNobody signs up for these, so there is no customer list to point at. What there is, is people using them: more than five thousand visits so far, across the tools. These three are a good place to start.
Plenty of sites claim everything happens in your browser. It is worth being precise, because for two of our tools it is not quite true.
Tools you give a file to read it on your own machine. The PDF extractor, the PowerPoint extractor, the PDF locker and unlocker and Grid Maker all work on the copy in your browser tab. Nothing is uploaded to us, we never receive the file, and closing the tab discards it. That is also why they keep working with your connection off.
Tools you give a web address to cannot work that way. A browser is not allowed to fetch another site's pages directly, so the Website Content Extractor and the Metadata Extractor send the address you enter to a small function on our own server, which fetches that one page and hands it back to your browser. Nothing is stored and no third party sees the address. Use them for public pages, and not for anything behind a login or otherwise private.
The full detail, including what our analytics and advertising do and do not collect, is in the privacy policy.
Four rules we hold ourselves to. They are specific on purpose, so you can check whether we are keeping them rather than take our word for it.
Every tool on the site works without signing up, without an email address and without a usage cap. There is no paid tier held back behind a limit.
Two of our tools send a web address to our server so it can fetch the page, so we say that plainly above instead of claiming everything is local. Where a tool has a limit, its page tells you before you use it.
You will not find a percentage or an average time saved anywhere in our writing, because we have not run studies that would make those figures honest. Arithmetic on your own numbers is offered instead.
Messages go to a real inbox and get read by someone who works on the tools. There is no ticket queue and no chat widget pretending somebody is standing by.
The articles are the same kind of thing as the tools. Each one answers a question somebody actually asks, gives a method they can follow, and says where that method stops working.
We do not publish invented statistics. Where an article offers a number, it is arithmetic you do on your own measurements, and where a rule of thumb is a judgement call we label it as one. Several pieces end with a short note explaining what the advice is based on and what it is not. When we get something wrong, the article is corrected rather than quietly replaced.
Where an article recommends one of our tools, it is because that tool does the step being described. The rest of the method works without it.
Read the Guides