It should be clear that I am looking at TTRPG rules. Rules generally would be far too broad a question for me, and as you say there has been much thought on that already.
So, here's a rule in Apocalypse World - it is written in a way so as to speak to the MC, and so uses the second person to refer to the GM:
The players’ job is to say what their characters say and undertake to do, first and exclusively; to say what their characters think, feel and remember, also exclusively; and to answer your questions about their characters’ lives and surroundings.
This rules establishes a normative standard for play - a set of exclusive permissions conferred on the players (to say what their PCs say, undertake to do, think, feel, and remember), and also obligations imposed on the players (to answer questions).
It doesn't take the form of
description is matched to norm/rule that yield consequence. It is about conferring permissions and obligations on participants.
Here is a rule, from 4e D&D, that superficially
does have that form: if a character falls into a pit, then (everything else being equal) they land prone. But the problem with that so-called rule is that it is not a rule at all! It doesn't state any normative standard. Hence, the better view is that my statement of the 4e rule is incomplete. The true 4e rule is that:
if the participants agree that a character falls into a pit,
then they are obliged (if everything else is equal) to agree that the character has landed prone. When stated correctly, as a normative standard, we see that this rule, too, is about conferring obligations on participants.
Like all game rules, RPG rules establish normative standards to which participants voluntarily agree to hold themselves - or, to put it another way, standards by which they voluntarily agree to be bound. Sometimes we state the rules in an elliptical fashion - eg
a queen in chess can move any number of squares but the correct statement of the rule will make clear what the standard is - eg
a player who is making a move in chess may move their queen any number of squares.
To reiterate: Working out the general form of RPG rules is not hard. Like other game rules, their general form is to confer permissions in respect of, or establish prohibitions on, the conduct of the participants in their capacity as players of the game.
(There is an alternative way of stating the rules of chess, in a quasi-mathematical fashion as constituting a set of possible game-states. In this form of stating them, normative standards that confer permissions instead become statements of possibility that underpin the construction of the set of possible game-states.
Perhaps something like this is possible for a very simple D&D combat, or even a very simple dungeon crawl. It's not possible for a 4e skill challenge, or for the play of AW. Hence the approach to RPG rules as stating normative standards for participants is the more fundamental. Which is Vincent Baker's point when
he says "So look, you! Mechanics might model the stuff of the game world, that's another topic, but they don't exist to do so. They exist to ease and constrain real-world social negotiation between the players at the table. That's their sole and crucial function.")