Shipping seven agent skills is a claim: hand a coding agent this tooling and it builds better bestax apps than it would cold. Claims want numbers. So we grade cold agents against a frozen rubric, feed every finding back into the skills, and keep the receipts. This post is the receipts.
A coding agent can't one-shot a library its training data barely saw. Ask one for a dashboard in an under-represented library and you get one of two failures: it quietly reaches for the incumbent it has seen a million times, or it stays loyal and invents props that don't exist. bestax's answer is to make the real, current API machine-legible so one shot is enough, and this post is everything that ships today.