Each learning here comes from repeated trial and error. A prompt fails, the cause is found, and the fix is written down. The aim is to save the next attempt from the same mistake.