---
title: Create a test
description: "Create your first test, run it on an agent, and read the result. Then pick the guide for your channel: web, API, smartphone, or USSD."
sidebarTitle: Create a test
lastUpdated: "2026-09-23"
---

> **For AI agents:** the complete documentation index is at [llms.txt](/llms.txt). Append `.md` to any page URL for its markdown version.

Every test in Kapptivate follows the same loop: author a test, run it on an agent, read the result. This page shows that loop, then points you to a channel-specific guide so you can start with the kind of test you care about.

## The loop

<Steps>
  <Step>
    **Create a test.** Open the **Tests** section for your product and click **+ Create new test**. Pick an authoring experience and a channel, then name the test. See [Creating a test](/tests/create) for the full setup flow.
  </Step>
  <Step>
    **Add steps.** In the [test builder](/tests/builder), record interactions on a live device or add steps manually, organized into step groups. Save the test.
  </Step>
  <Step>
    **Run it.** Click **Run** on the test's row, then review the [run confirmation](/executions/run): adjust the device, variables, and options, and launch.
  </Step>
  <Step>
    **Read the result.** Open the execution under **Executions** to see status, per-step screenshots, timing, and variables.
  </Step>
</Steps>

## Pick your channel

<Columns cols={2}>
  <Card title="Web" icon="globe" href="/agents-devices/web-api">

    Automate a browser test against a website or web app

</Card>
  <Card title="API" icon="code" href="/tests/actions/api/overview">

    Validate HTTP endpoints with methods, headers, bodies, and response assertions

</Card>
  <Card title="Smartphone" icon="mobile-screen" href="/agents-devices/smartphone">

    Test a native Android or iOS app on a real device

</Card>
  <Card title="USSD" icon="sim-card" href="/agents-devices/sim">

    Run USSD and SMS flows on a real SIM card

</Card>
</Columns>

## Go further

Once you've run a test:

- **Organize tests** into [collections](/tests/overview) by feature or team.
- **Parameterize tests** with [input variables](/tests/configurations) to run the same test with different data.
- **Run on a schedule** by [creating a monitor](/monitoring/create) that runs continuously and alerts you to failures.
- **Integrate with CI/CD** using [API keys](/administration/api-keys) to trigger runs from your pipeline.

## What's next?

<Columns cols={2}>
  <Card title="Navigating" icon="compass" href="/concepts/interface">

    Find your way around the sidebar, workspaces, and products

</Card>
  <Card title="Core concepts" icon="lightbulb" href="/concepts/core-concepts">

    Understand workspaces, products, tests, executions, and monitors

</Card>
</Columns>