The character in the fiction
did interpret the runes: they studied them, and worked out what they said.
At the table it's true that the GM didn't just tell the player what the runes say: rather, a resolution process was used. But if you are agreeing that this matters to the diegetic nature of a resolution process, then I think you are agreeing with
@Hussar: diegesis concerns the relationship between events the audience experiences, and events in the fiction.
Here is how Oxford Languages, via Google, defines "diegetic":
(of sound in a film, television programme, etc.) occurring within the context of the story and able to be heard by the characters.
The general concept pertains to a thing that is experienced by those who are observing/experiencing the fiction (the audience) that also occurs within the fiction.
@Hussar's example has been music in a movie that is coming from a radio. That contrasts with music that is not part of the fiction - eg most of the music in Star Wars.
I think the best example of a diegetic RPG resolution process that's been offered is Hussar's example of the map, or the puzzle square, being handed by the GM to the players. This is something the "audience" - the players - are experiencing which is the same as what the characters are experiencing (they, too, are studying the map or the puzzle). Rolling dice can never be diegetic in any literal sense, as the players experience the dice but the characters don't. Because of this, I'm not 100% sure how the metaphoric extension of "diegetic" to RPG mechanics is supposed to work, but clearly it has to pertain to the relationship between the game participants (and their real world experiences) and the events in the fiction (and whether those real world experiences are in some fashion part of the fiction).
Clearly, the resolution process for the reading of the strange runes isn't "diegetic", in that (a) the process of authoring the fiction, that works by having the participants accept the output of a resolution process, is (b) not something, nor a correlate or representation of anything, that the characters in the fiction experience. But then neither is a GM's decision that, as a result of a failed climb check, a PC falls or their rope is cut by a sharp rock or whatever: that decision, which is an event in the real world, is not something that correlates to or represents anything in the fiction. Which is at least part of what I take to be
@Hussar's point.
My post was not about in-fiction causal dependencies; just about the dice rolls. In particular, that bundling two things - one dependent on a skill and one independent of a skill - into a single roll adjusted by the skill is not an unusual thing in a RPG.
I don't assert that it is a simulationist mechanic. But by some measures - eg the idea that simulationist play is about a certain sort of experience - it is consistent with simulationism, that is, the focusing of attention on a particular bit of the fiction.
On the "supplanting": I go back to the spear case. Once the spear is in flight and on target (as determined by the skill of the thrower), the chance of the intended victim to dodge is not a function of the thrower's skill. It depends on their reactions/reflexes and their speed/agility. (I'm deliberately choosing a spear rather than a gun shot for this very reason - because the spear can be dodged once in flight. I'm not sure either way about arrows.) Yet in most versions of D&D, the two things - accuracy of throw, and success of dodge - are bundled into a single roll.
It also produces some variations from other systems. Let's suppose that skill bonuses are a (rough) measure of skill. So +8 to hit is twice as skilled as +4 to hit, etc. In D&D, if the bonus to hit (added to the roll) and the bonus to dodge (subtracted from the roll) are equal and are both doubled, the chance to hit remains the same. Whereas consider RQ: a 40% bonus to attack and to dodge produces a 2/5 * 3/5 = 6/25 chance to hit. Double those to 80%, and the chance to hit is 4/5 * 1/5 = 4/25. That is, the chance of a hit has reduced by a third (from 6 to 4 chances in 25). This is why RQ can tend to produce whiffy combat. But it also shows that the idea of a simple correlation between the numerical bonus and "how good" a character is at a thing doesn't work - it depends on how the bonus is factored into the details of a resolution system.