Getting Started

1. Create your first project

Go to Setup > Projects > Create Project in the sidebar.

A project is an organizational bucket that groups your journeys, agents, schedules, and other resources together. Projects have no behavior of their own. They exist so you can scope your testing in whatever way makes sense for your team. Two common approaches:

• By environment. Create separate projects for development, staging, and production. This lets you run different test suites against each environment and track their health independently.
• By service or application. If you have multiple web applications or microservices with their own frontends, give each one its own project.

Give your project a good name and description, then click Create Project. You can always come back and update these later. Every agent you create will belong to a project, and the project health board gives you an at-a-glance view of all your journeys and their current pass/fail state.

Want to learn more? See the Projects API reference.

2. Create your first journey

Go to User Journeys > Create Journey in the sidebar.

A journey is a re-runnable prompt. Instead of creating a new agent from scratch every time, you define a journey once and run it as many times as you want. Each run creates a new agent linked back to the journey, so you build up a history of results for the same test over time. This is how you track whether something is working or broken.

You will see three fields:

• Project. Select the project you just created.
• Where should we test? Enter the URL of the page you want to test. This is where the agent will start its browser session.
• Journey description. Describe what you want to test in plain language. For example: "Sign up for a new account, verify the confirmation page shows a welcome message." The agent will receive this as its prompt and figure out how to accomplish it.

When you click generate, we will create a journey for you and show you a preview. You can edit the generated journey if you want to tweak anything. Once you are happy with it, click create and the journey is saved to your project.

You can also add checkpoints to your journey. These are named milestones that you expect the agent to reach during execution (like "logged in", "added item to cart"). They are evaluated after the agent finishes by analyzing what it actually did, giving you structured visibility into how far through the flow it got.

Want to learn more? See the Journeys API reference for details on checkpoints, success determination, and persona/role resolution.

3. Set up a schedule

Go to Schedules > Create Schedule in the sidebar.

A schedule ties a set of journeys to a recurring cadence. You pick a project, choose which journeys to include (or add all of them), and set when and how often they should run. The system handles the rest, including rotating through journeys if your daily agent budget does not allow running all of them in a single tick.

Schedules are budget-aware. Each journey in a schedule has a priority weight, and the system uses a scoring algorithm that picks the stalest (longest since last run) and highest-priority journeys first. This means even with a limited budget, all your journeys get covered over time.

You can also enable an email digest so your team gets a summary of the results after each run. Configure it to send always or only when failures occur, and set the recipient list.

Want to learn more? See the Schedules API reference for details on coverage windows, budget enforcement, and digest configuration.

4. Add more journeys

Now that you have the basics in place, start adding more journeys to cover the important flows in your application. Follow the same steps as in section 2: go to User Journeys, create a new journey, pick your project, describe what to test, and generate.

When you add a new journey, make sure to include it in the schedule you set up in section 3. If your schedule is configured to run all journeys in the project it will pick up new ones automatically. Otherwise, edit the schedule and add the new journey to the list.

As you build up more journeys, consider setting up personas if your application requires login. A persona carries its own credentials and email inbox, so agents can authenticate automatically. You can also assign roles to shape how the agent behaves during execution, for example testing as a power user versus a first-time visitor.

Want to learn more? See the Personas and Roles API references.

5. Further reading

Once you are comfortable with the basics, there is more you can do:

• Integrations Connect Linear, GitHub, and Slack to your testing workflow.
• Use Cases Real-world examples of how teams are using Aiqaramba.
• API Reference Full documentation for every endpoint, including agents, discoveries, memories, and files.