/
No results found

Core Concepts

Four concepts cover almost everything you do in Aiqaramba: projects, tests, findings, and test plans. This page defines each one and then walks through how they work together in a basic setup.

The Building Blocks

  • Project. The container for one application you are testing. Everything else lives inside a project. Most teams create one project per application, or one per environment such as staging and production.
  • Test. A single check, described in plain language rather than code. You write the goal, for example "sign up for a new account and verify the welcome page loads," and you can run it as often as you like. Each run sends an agent to a real browser to carry out the steps and report back what happened.
  • Finding. Something an agent noticed during a run that you should know about: a bug, a broken flow, or a smaller observation. Each finding carries a severity (critical, high, medium, low), screenshots of what went wrong, and steps to reproduce. Findings are how a test run turns into something you can act on. When integrations are connected, a finding can automatically open an issue in GitHub or Linear.
  • Test Plan. Several tests chained into an ordered sequence, where a step only runs once the step before it passes. Use a test plan when the order matters, such as "create an account, then log in with that account, then complete a purchase." See the Test Plans guide for details.

How They Fit Together

These four pieces stack on top of each other. Here is what a basic setup looks like and how it produces results you can act on.

Start with a project for the application you want to test. Inside it, write a handful of tests, one for each thing that matters: "log in with the test account," "add an item to the cart," "submit the contact form." Keep each test focused on a single outcome so that when one fails you know exactly what broke.

When some checks only make sense in order, group them into a test plan. Signup is the classic example: register an account, then log in with it, then load the dashboard. Each step runs only after the previous one passes, so the plan stops at the first thing that is actually broken instead of reporting a cascade of failures.

Every time a test or test plan runs, an agent carries it out in a real browser. If it hits something wrong, it records a finding with a severity, screenshots, and the steps to reproduce. A green run with no findings means the flow still works; a run with findings tells you what to fix and where. That loop, tests and test plans in, findings out, is the core of Aiqaramba. Everything else builds on it.