Your words are all it has

The model does not know your repo. It does not remember yesterday, and it cannot see the file you didn't attach. Its entire universe is the text sitting in front of it right now.

That is the deepest rule of prompting, and it isn't mystical: you are deciding what exists. Whatever you put in front of it is the world; whatever you leave out never happened.

So the question behind every prompt is not "how do I phrase this?" but "what does the model need in order to get this right?" This chapter is the first half of that answer — the words themselves. The next one is the harder half.