SanityCheck
AI agents click through your app in a real browser to test it, from a plain-English sentence
SanityCheck spins up real browsers in the cloud where an AI agent follows a plain-English test description, like adding an item to a cart and checking the confirmation, then reports back with a full replay of every action, console log, and network request. Test credentials are encrypted and filled by direct tool calls, so the agent never actually reads the password. It can trigger runs automatically on every GitHub, GitLab, or Vercel deploy, and gives 15 free agent test runs a month with no credit card.
What holds up
- +Agent never sees the actual password — credentials are filled by tool calls, not the model
- +Every run has hard step and time budgets plus "never do" rules enforced by the platform
- +15 free test runs a month with no credit card and no test code to write
Mind the limits
- −Tests describe one outcome per run — complex multi-step flows may need several separate tests
- −Deploy-triggered testing needs GitHub, GitLab, or Vercel — no support for other CI shown
Picked here? Add the badge
Put it on your site. It links back to this page. Free for every listed product.
<a href="https://stillworks.dev/products/p/sanitycheck/"><img src="https://stillworks.dev/badge/sanitycheck.svg" alt="Picked by StillWorks" width="230" height="54"></a>