Explanation
Background reading for when you want to understand how b8e behaves, away from the keyboard. For
steps and flags, go to the how-to guides and the
reference.
- About the two scanning models: static source vs live DOM — which command answers which question, and what each one can and cannot see.
- About components the scanner can and cannot see — how your wrappers get traced to the elements rules are written about, and what to do when one cannot be.
- About how findings are ordered — severity first, reach as a tiebreak, and why the order can differ between your laptop and CI.
- About TalkBack audits — how an Android audit drives the real TalkBack, what the agent can do on the device, and which device settings it borrows.
- About VoiceOver audits — how an iOS audit drives the real VoiceOver on a simulator, what the agent can do there, and what a run changes on the simulator.
- About Agentic QA — how a QA step is performed and judged, what each outcome means, and why only a failed step is a Finding.
- About usage telemetry — exactly what one event contains, what it never contains, and the two ways to switch it off.