Bilingual runs
In setup you tick the languages your app has and which one is its main language. From then on every journey runs once per language. The two runs are queued together; where a journey stands is the worse of the two, so a failure in Arabic shows red even when English passed.
Burhan writes the journeys in English, with Arabic titles and step notes next to them, and writes its run notes in Arabic for the Arabic run. The whole dashboard is in Arabic too, if that's how your team reads.
What it catches
The Arabic run isn't a translation check. Burhan does the same steps and expects the same results, so it finds the places where the Arabic version behaves differently or reads wrong:
- English text left in the Arabic screen: a button, an error message, a status label, an empty state.
- Layout that didn't flip: a back arrow pointing the wrong way, a progress bar filling from the wrong side, a list that's cut off at the edge.
- Mixed direction: a price or phone number split around Arabic text, a date reading backwards, an SAR amount with the sign on the wrong side.
- Text that doesn't fit: Arabic is often wider, so labels truncate, buttons wrap, and tabs overflow.
- Behaviour that differs: a form that validates in English but not in Arabic, a search that finds Latin but not Arabic input, a language switch that drops the user's place.
How a red reads
A red in the Arabic run says the step, what Burhan expected, what it saw (in Arabic, with the English next to it), and the screenshot and video at that moment. If it's how your app is meant to work, mark it "Not a bug" and say why; Burhan judges with that from the next run.
What's not covered
Burhan reports what a user would notice. It doesn't grade translation quality or tone, and it doesn't compare the two versions pixel for pixel. Numerals follow your app: if you show Arabic-Indic digits, Burhan reads them as digits.