Glossary
Naur's shadow
The mental model that lives inside the original programmer's head. Programs decay because that theory walks out the door — and AIs can't reconstitute it without help.
Definition
Naur's shadow comes from Peter Naur's 1985 essay Programming as Theory Building, the most important paper in operator history that nobody operating reads.
Naur's argument: a program is not the source code. The program is the *theory* in the original programmer's head — what the code is *for*, why this design and not that one, which constraints matter. When the programmer leaves, the theory leaves with them. The code rots not because it changes but because nobody remembers *why* it's the way it is.
This is the core problem AI agents face today. They can read your code, but they can't read your shadow. The Command Kit exists to externalize the shadow — to write down the theory so an AI can render it back.
We use "Naur's shadow" as shorthand for this missing context. Every time an AI does the wrong thing because it doesn't know your business, that's the shadow it can't see.