Cofactor reads what your team has already written, writes the tests, and runs them against your app.
Cofactor uses the documents your team already has, and checks them against the app that is actually running.
Specs, PRDs and acceptance criteria describe the behaviour you agreed to ship.
Tickets show what the team is shipping this week and what broke last month.
Your help centre and internal runbooks are where the real workflows are written down.
Cofactor explores every page it can reach, so the suite matches what actually exists.
Anyone on the team can read a test, follow it and change it. There are no selectors, page objects or frameworks to learn.
The flows that cover what you changed run as a required check. It takes minutes because it skips the rest of the suite.
Every flow runs for every persona on a schedule, and the report is ready when your team starts work.
Start a session from Claude or Cursor and drive a real browser locally while you build.
Where do the tests come from?
Cofactor writes them from what your team already has, like product requirements, tickets, specs and help docs, and from exploring the running app. It proposes the flows and you approve them.
Do I have to write selectors?
No. A flow is a list of steps in plain English. Cofactor finds each element when the step runs and finds it again if the interface moves.
What happens when the product changes on purpose?
Cofactor notices the change and asks whether the test should be updated. If you say yes, it updates the test for you.
Can we bring the suite we already have?
Yes. Cofactor runs your existing suite as it is, in parallel on our machines, and diagnoses every failure. You get the same reports without rewriting anything first.
Join the waitlist, or talk to us about the suite you maintain today.