What problem does this address?

In addition to other test types, we rely upon E2E tests for critical flows related to native OS features. The current native mobile editor E2E tests provide a good foundation for our needs. However, there is opportunity to increase the tooling robustness, refactor to make it more approachable, and remove dead code/unused abstractions now that we prioritize integration tests.

What is your proposed solution?

We could work on the following tasks to improve the robustness and approachability of the E2E tests.

• Clearing Metro caches before bundling for E2E tests, as it can cause cryptic failures between test runs after running npm install - test: Improve E2E test consistency #55554
• Update the device-tests:local command to launch the Android emulator via the emulator CLI. #55567
• refactor: Remove unused E2E test helpers #55711
• test: E2E test setup script creates iOS simulators #55716
• test: E2E test setup script creates Android emulator #55898
• Refactor E2E test EditorPage and utils to support applying changes in watch mode
• Reassess impact and necessity of artificial delays, durations, etc
• Remove low-value, single-use E2E test abstractions considered unnecessary indirection