Assign a Live user or Test user on the test's Start node. The assigned user must match the database for the selected branch.
Running and reviewing tests
Run a test against your Bubble app, then review each step, screenshot and result to understand what passed or failed.
Run one test
You choose the Bubble branch and AI model before each run.
Open the test and select Run.
Choose Live, Test or a custom branch.
Choose the model.
Select Start run.
A Live run uses the assigned Live user. Test and custom-branch runs use the assigned Test user. The run can start without one, but the agent will not have a pre-configured login.
Tests can perform real actions in the selected database. Check the branch and assigned user before starting a run against Live.
Run several tests
Run a folder when you want to check several related flows together.
Buildprint starts the runnable tests as separate runs and combines their status in History. Reusable components are included through their tests and do not run by themselves.
Review a result
A completed result shows the overall status, branch, database, model, assigned user and outcome of every step.
Passed means the agent completed the step successfully.
Warning means the run continued, but the step found a problem.
Error means the step failed.
Skipped means the step was not needed, such as the unused side of a condition.
Select a step to read what the agent did and inspect its evidence. Screenshots show the app at that point in the run. Attached artifacts appear with the result when available.
Use History
History keeps earlier runs so you can compare results and repeat a test.
Open a run to review its steps and evidence. Use rerun when you want to run the same test again, then confirm the current branch and model.
Monitor regressions
The Dashboard highlights recent runs and regressions: tests that passed before but now have a worse result.
Use the time range and status filter to focus the Dashboard. Open a recent run to inspect the change that caused it to fail.
Run tests after deployment
Enable Automatically run tests on deploy on the Dashboard to check your app after a Bubble deployment.
Review automatic runs in the same Dashboard and History views as manual runs.
Fix run problems
The run starts without a logged-in user
The run fails before the first step
Check that the selected branch has an app URL, the test user is active, and the selected model is available through one of your connected providers.
A step fails even though the app looks correct
Make the step more specific. Include the exact action, expected result and any text or value the agent should check, then rerun the test.
A run stops after one failed step
The step uses Stop on failure. Change it to Continue on failure if later steps can still provide useful results.
Was this helpful?
Was this helpful?