Hi Product Hunt 👋 I'm Russell, founder of SanityCheck. The idea is super simple. Test your apps like an actual user. Here's why I built it. On another project of mine, the database credentials were rotated and never updated in the app. The code was fine. The logic was fine. The homepage was static, so it returned a 200. Every automated test passed but moment I opened a page that needed data, it failed. If anyone had just clicked through the app, they'd have caught it in ten seconds. You (or your agents) can write all the automated tests you want. A lot of the time that's still not enough or doesn’t cover every edge case. Sometimes you just to actually use the thing. SanityCheck does exactly that. Describe what to test. That’s it. "Go to tokenrouter.io, click through every page, log in, and make sure the console works." That's the whole test. An AI agent opens a real browser and does it: clicks, navigates, scrolls, reads. Anything you can do, it can do. You can watch it live, and every run is recorded with video, console, and network so you can see exactly what it saw. No test code. No flaky framework. No CI config. - Store login credentials once, encrypted, and the agent uses them without the model ever seeing them - Run tests on every deploy from GitHub, GitLab, or Vercel - Or schedule a daily run that just makes sure your app is still up and working A year ago browser agents weren't reliable enough for this. Now they are, and I think this is what QA looks like for small teams going forward. Free tier is 15 runs a month, no card. Give it a try and let me know what you think!

SanityCheck

AI agents that QA-test your web app in real browsers.

AI agents that QA-test your web app in real browsers.

SanityCheck

Previous SanityCheck Launches

Previous SanityCheck Launches