Skip to content

Running Tests

Overview

You can run tests in VibeTesting in two execution modes:

  • Exploration — the agent freely explores your app within defined focus areas
  • Guided — you provide Tests Guidelines (flow goals or step-by-step), and the agent stays on-path

All modes run in the cloud and produce a results summary. You can also enable an email report that includes screenshots.

Quick setup flow

  1. Sign up / log in
  2. Under Application, enter your application URL
  3. Open the Agent page
  4. Start a run (Exploration or Guided)

Choose an execution mode

Running-tests screenshot

Example: Starting an Exploration run - the agent asks for login, focus areas, number of flows, steps, and whether to send a final report by email.

Use exploration when you want:

  • Fast sanity validation
  • Broad coverage without writing steps
  • Automatic discovery of obvious UI/flow issues

Continue: Running Tests → Exploration Mode

Guided mode

Use guided mode when you want:

  • To test specific user flows described in natural language
  • The agent to follow your intent strictly without free exploration
  • Goal-oriented testing or step-by-step testing using Tests Guidelines

Continue: Running Tests → Guided Mode

Suites and reruns

  • Group tests into suites
  • Re-run by suite name any time — from the agent chat or via the External API
  • Use reruns to validate fixes and catch regressions

Example suite names:

  • Checkout process
  • Adding new items
  • Sanity before release

Real-time logs

Execution logs are streamed in real time while tests run. See Real-Time Logs for details.

Batched execution

Run tests in parallel batches to shorten feedback loops. See Batched Execution for details.

Coming soon

  • Scheduling runs ahead of time
  • MCP support
  • Additional user actions (drag and drop, multiple tabs/windows)