Method
End-to-end testing without writing scripts
End-to-end scripts are worth having and expensive to keep. Here is what you can check without writing any, what that costs you in precision, and how teams combine the two.
Can I do end-to-end testing without writing test scripts?
Yes, for checking that users can get through: AI agents take a task in plain words, use your app in real browsers, and report what broke and where they got stuck. Mimiq does this with no selectors and nothing to maintain when the page changes. You give up exact repeatability, so keep a few scripted tests for behavior that must never change.
The maintenance tax
A scripted suite starts as ten tests and becomes a second codebase. Every redesign breaks selectors, every new step needs a new test, and flaky tests get skipped until nobody trusts the suite. Small teams often stop writing them at all and test by hand before releases.
- Tasks in plain words, no selectors.
- Nothing breaks when the page is redesigned.
- Agents notice confusion, not just failures.
- Browser records: errors, failed and slow requests.
- Not deterministic: keep scripts for exact checks.
How it works
- 01
Give it a link and a task
Paste a public address, such as a preview deployment or staging, and say in plain words what a user would try, like "Sign up and start using it". No test scripts and no selectors.
- 02
Pick how many agents
1, 3, 5, 10 or 20 agents, 5 by default. Each acts as a different one of your users and gets its own real browser for up to 40 steps. About 5 minutes for up to 3 agents, about 5 to 15 minutes for 5, and about 15 to 30 minutes for 20. Agents take turns on shared browsers, so it can take longer when other runs are going, and you can close the tab and come back.
- 03
Read what they hit
The report shows what broke, what the browser caught on your site (JavaScript errors, requests that failed or answered 4xx or 5xx, and requests slower than 3 seconds), where each agent got stuck or gave up and why in its own words, with the step and a screenshot, and the problems ranked by how many agents hit them, each with a suggested fix.
- 04
Fix it and run the same agents again
Rerun on the fixed build and see whether the problem is gone for everyone who hit it. Or let your coding agent do the loop through the MCP server, or let the GitHub Action do it on every preview deploy.
How teams combine them
Keep a handful of scripted tests for money and permissions: the checkout total, who can see what. Point agents at everything a user goes through: signup, onboarding, settings, new features. Run both on pull requests.
Agents are simulated users. They find problems fast and on every build; they do not forecast how many real people will convert, they are not a deterministic regression suite, and they do not install native mobile apps.
Scripts and agents
| Scripted end-to-end tests | Mimiq agents | |
|---|---|---|
| How a test is written | Code: selectors, actions, assertions | A task in plain words |
| When the page changes | Selectors break and need fixing | Agents read the new page |
| Same result every run | Yes | No: different users take different paths |
| Finds confusing wording or layout | No | Yes, in the agents' own words |
| Finds paths nobody planned | No | Often |
| Records browser errors and failed requests | If you assert on them | Always, on the step they happened |
| Best for | Behavior that must never change | Whether new users can get through |
Questions
Can Mimiq export Playwright tests?
No. Mimiq keeps no test code. If you want generated scripts you own, see the comparison with Playwright and with tools that export tests.
How precise are the results?
Browser records (a 500, a JavaScript error) are exact. Where an agent got stuck is real for that agent; how common it is shows in how many agents hit it.
Keep reading
Let the agents find what breaks first.
Your first test is free: sign in with Google and it starts right away. Last checked 2026-10-22.