Run tests and view results

Run

• Run all tests: ▶ Run All at the bottom of the screen (⇧⌘⏎ / on Windows, Shift+Ctrl+Enter)
• Run the selected test: ⌘⏎ (on Windows, Ctrl+Enter)
• Stop a run: ⌘.. The list shows Stopped.

By default, multiple tests run 3 at a time in parallel. You can change the number of concurrent runs to 1–8 in Settings → Run. Tests that must run alone, such as those that change shared state, run one at a time so they don't overlap with other tests.

During a run

Each row in the list shows Running / Needs Action / Queued / Passed / Failed / Stopped / Not Run. A small window in the corner of the screen shows the running browser, and you can enlarge it with ⤢. When multiple tests run in parallel, the small window shows one of them.

When a person's input is needed, such as a two-step verification code, the test stops with Needs Action. → Stopped at "Needs Action"

When you're not looking at the app, OS notifications appear when a test needs action and when Run All finishes. You can turn notifications on or off in Settings → Notifications.

Run at set times

In Settings → Schedule, choose a repeat pattern (daily, weekdays, selected days of the week, every few hours) and a time to run all tests or selected tests.

• It only runs while the project is open in echo. Runs that came due while echo was closed or the PC was asleep run once the next time you open it.
• If another run is in progress, it runs after that one finishes.
• Since no one is assumed to be present, it doesn't ask you to log in. Tests for accounts without a saved login state are skipped, and tests whose login has expired fail.
• The settings apply to this PC only. Runs appear as Scheduled in the run history.

View results

Click a row in the list to open its details on the right, where the steps and the list of Checks are laid out like a storyboard, with screenshots from the latest run.

On the result screen, you can play the recording and click a step to jump to the screen at that moment. Past results for that test are listed at the top, newest first; click one to switch to that run. Open Trace opens Playwright's Trace Viewer, where you can follow network traffic and even the elements that were clicked in detail.

You can also have AI fix a failed test right there.