test-browser

End-to-end browser tests can target affected pages by identifying changed files from a Git diff, mapping them to routes, and verifying server status before testing. The agent-browser CLI executes these scoped tests with snapshotting and interaction verification.

Automated regression testing for pull requests is achieved by determining scope through changed files, mapping them to testable routes, and running browser tests exclusively on those affected paths to catch regressions efficiently.

When automated browser testing encounters failures during interaction verification, it handles them by documenting the failure, allowing debugging, creating todos for follow-up, or skipping the test to maintain workflow continuity.

Yes, this browser testing approach requires no external dependencies, relying solely on the agent-browser CLI to execute end-to-end tests, capture snapshots, and verify interactions directly within your local environment.

Visual regression testing requires mapping changed files to specific routes because the testing scope is determined by Git diffs; this ensures only pages affected by current code changes are snapshotted and verified.

Automated E2E testing includes server status verification as a prerequisite step; it checks the server availability before attempting to execute browser tests on the mapped routes to ensure valid results.