pinned
projectopen source · MIT · Evo Tech
The agent writes the test. You own it.
Proba is a QA workbench. An agent drives a web app, API or database through MCP, and every action becomes a test you can read, edit and run again. It runs fully local. At Evo Tech we use it in QA for our internal systems: for evidence, and for tests that run the same way every time.
One test, three views
Change one place and watch the others follow
Feature: Sign in
Scenario: User signs in with email
Given I open "/login"
When I fill "Email" with {{account.client.email}}
And I press the button "Sign in"
Then I see the heading "Dashboard" One behaviour, three ways to read it.
The loop
Record, finalize, replay, repair
Replay
The test runs again whenever you want. If it fails, Proba files a task on the board with the exact step.
Why it is different
The agent does not start from zero every time
The tests are yoursYou can edit any of the three views. The change goes through the shared form and reaches the other two.
Memory across sessionsWhat the agent learns about the app stays. Next time it does not rediscover everything.
Stable locatorsRole, text, label or test id only. Brittle positional selectors are rejected.
Web, API and databaseOne spine for all three, plus visual, layout and accessibility checks.