You can try looking for old VT-100 (or the later VT-* family), IBM 3270 and successors, or Wyse terminal manuals. Maybe on eBay or Amazon. At their simplest, terminal UIs just spit out text that displays wherever the cursor happens to sit, usually at the bottom left so it works like an old printing terminal (Teletype). For more sophisticated terminal UIs the application can send control codes and escape sequences to position the cursor, replace or redraw portions of the screen, draw line graphics, etc. Each terminal manufacturer made their own codes with the IBM and DEC (VT series) widely emulated.
I'm sure you can find online documentation for the functions and escape sequences of all of the common terminals. For example: https://vt100.net/docs/vt100-ug/chapter3.html
You might also look at the code for open source terminal emulators.
From the documentation you can probably infer how terminal UIs work. I did a lot of terminal UIs in the past. In general the application on the back-end (mainframes and minicomputers in the old days) maintained an 80x24 (or larger) array of characters and metadata, application code made changes to that screen buffer, then a layer in between translated changes to the correct escape codes and sent that to the hardware terminal. I'm simplifying but that's the general idea.