Exploration Mode¶

What exploration is best for¶

Fast sanity validation
Finding unexpected UI/flow issues
Learning how users can get "stuck" on real screens

What you configure¶

When you type "Test my app", the agent will ask about:

Whether login is needed (open app or first-party username/password — see Quickstart)
Focus areas (pages/features to explore)
Optional General Instructions — guidelines / known issues (up to ~1000 words)
Number of flows
Max steps per flow
Optional email report

Exploration-mode screenshot

How exploration behaves¶

The agent tries different safe actions ("sampling") to increase coverage
It respects your focus areas and guidelines
It flags issues with severity and context

Batched exploration¶

When using batched execution, exploration batches can cover different themes or focus areas, increasing coverage across your application in a single run.

Monitoring progress¶

Exploration runs in the cloud. You can monitor:

Current flow and step
Actions taken
Real-time logs via the logs dashboard
Intermediate status (the final email report includes full details and issue screenshots)