The scenario is the asset; runners are swappable. Define mobile app journeys once, run them through any agent, device, or profiler, and preserve stable evidence that tells agents and humans whether behavior improved, regressed, or can't be trusted yet.