Run code you did not write and did not read, choosing each confinement because of the specific thing it prevents, and be able to say exactly what an attacker who fully controls that code can still reach
Letting a Model Call Your Code Safely

Executing untrusted code written by a model, derived from what code can actually do to a machine and what each boundary genuinely stops.
8 lessons, written and corrected before you arrived. Reading them here needs no account. The first reads the whole way through; the others open and then stop, because a page nobody owns cannot tell who is reading it. Starting the course gives you your own copy, where every idea has problems standing under it and you can ask about any sentence.