Sometimes CTF challenge writers don't want to hurt your head with a normal language like C or Python, and instead give you something completely random.
Sometimes it will be old programming languages, like FORTRAN or ADA. If you see a language that you don't recognize, look for something that appears to be the main function or a loop, basically any construct you recognize, and google it. If it is an important piece of the language, someone has mentioned it before and you'll at least know the language name. From there, you have to find some old documentation (or hopefully example code on Github) to copy. You'll have to find a compiler on Github to test them, and getting the entire toolchain to work is a real pain. These challenges are crazy hard, but it feels awesome when you get a punchcard program to compile.
Another classic is esoteric programming languages, which are defined by Wikipedia as " designed to test the boundaries of computer programming language design, as a proof of concept, as software art, as a hacking interface to another language (particularly functional programming or procedural programming languages), or as a joke." As most things are Turing Complete if you try hard enough (look it up if you don't know what that means, basically anything can be a programming language. Good for challenge developers, bad for you. For the most part though, they'll be using a well-known esoteric like brainfuck, or any of the other languages listed on this wiki page. Writing your own esoteric for a CTF is a lunatic move, so if you see something that doesn't match any of these formats, expect it to be a substitution version of an existing language, so look for syntactical similarities and throw it into an existing interpreter once you've translated it over.
Another really common challenge is using an alphabet from another language, especially fictional and scifi ones. For those, use a site named omniglot to get a comprehensive list with examples. As a fun note, you don't actually need to know the language to solve problems like this. As long as it is a standard 26 character alphabet, you can solve any made up language like a substitution cipher using standard cryptographic tools, just map each made up character to a real character and treat it like a substitution.
This is all to test your out of the box thinking and ability to learn quickly. It's terrible but they are actually kind of fun.
- Decrypt this: ++++++++++[>+>+++>+++++++>++++++++++<<<<-]>>>+++.<+++++++++.>>+++++++++.<<-------.>>++++++.----.+++..+++++++.<<++++++++++++.------------.>.<.>>---------------------.+++++++++++.-.<<+++++++.>>++++++.<<-------.>>--------.---.++.------.<<.>>+++++++++++++++.------------.+.++++++++++.<<.>>------------.------.++++++++++++.--------.<<++++++++++++++.