Short answers, pulled from the story.
A command-line interface lets a person control software by typing commands formatted as lines of text instead of using icons or menus. It is common in operating systems, programming language environments, and many applications. It is especially popular among programmers, system administrators, and people with visual disabilities who use refreshable Braille displays.
The first true command-line interpreter appeared with the Compatible Time-Sharing System, or CTSS, developed at MIT in 1961. It let a user type a command such as LOGIN or PRINT for a program called the supervisor to parse and execute.
Louis Pouzin coined the term shell in 1964 while working at the MIT Computation Center on the Multics operating system. He built a tool called RUNCOM for running command scripts with argument substitution. He then returned to France in 1965, after which Glenda Schroeder completed the first Multics shell.
Graphical user interfaces largely replaced the command-line interface because they let users act through icons and drop-down menus instead of memorizing commands. The shift began with the PARC approach used in the 1983 Apple Lisa and 1984 Apple Macintosh. Most IBM PC users, though, did not give up their COMMAND.COM shell until Windows 95 arrived in 1995.
Internal commands, also called built-in commands, are recognized and processed directly by the command-line interpreter. External commands run separate executable files that the interpreter searches for by matching names.
In Bash and other Unix shells, the prompt is controlled through a variable called $PS1, and zsh adds a right-side prompt with $RPROMPT. In DOS and Windows, a user can change the prompt with the PROMPT command or by editing the %PROMPT% environment variable. Typing PROMPT $P$G produces the standard C:\> style.