It seems like your "guidelines on how to apply the general mechanics in this situation" are the same as what I'm asking for: A rules framework for resolving mass combat. If it uses PC skills as a basis, that's fine by me.
<snip>
If it can be built out of the existing rules, all the better.
To get more specific about what I have in mind: the 4e DMG2 has guidelines on using money in a skill challenge. The suggestion is that expenditure equal to 10% of a magic item of the PC's level should be roughly equivalent to a successful secondary check (eg +2 to a primary check).
Now I personally don't think this is very well thought out - a consumable magic item of a PC's level costs 1/25 (=4%) of what a magic item of that level would cost, and frequently will contribute quite a bit more than the rough equivalence of a successful check; and rituals will also be closer in expense to those consumables and are suggested to be treated as automatic primary successes.
But anyway, by combinining this guideline with the rules for hirelings in MME, I can start to work out what sorts of contributions hiring an army should make to a skill challenge, in principle at least. The problem is I'm pretty sure that 10% figure wasn't thought through systematically (for the reasons I've stated) and I'm pretty sure no one thought about it when the set the costs for hirelings. Good guidelines would at a minimum (i) link this stuff together so that it all coheres well, and (ii) join the dots clearly rather than leave it as an exercise for the reader.
Although I see 4E's skill challenge "rules" as a way of saying, "If you want to build your own mechanics to handle a social encounter or some such, here's a starting point." A straight-up, by-the-book skill challenge is one of the most boring and pointless subsystems I've ever met in an RPG; so much so that I refuse to believe that was how it was meant to be used.
I'm close to certain that skill challenges, as a mechanic, were inspired by comparable mechanics in HeroWars/Quest and Maelstrom Storytelling (and probably other indie games that I don't know). The guidelines published by WotC are close to hopeless, but when you read the guidelines from the other games (i) you can see hints of them in what WotC said in the DMG, and (ii) you can (in my experience at least) get good results out of skill challenges.
I think the biggest failure of skill challenge presentation is in how samples are statted out, which appear to violate the stated rules. The rules for skill challenges state that the GM describes the fictional situation, the player describes how his/her PC engages that situation, the GM then specifies a skill check, the player then rolls it, and the GM then adjudicates that result in terms of its impact on the fiction. Rinse and repeat until either N successes or 3 failures and you have a skill challenge.
But the published skill challenges give the impression that the GM is to get the players just to roll checks, and the fiction is to be narrated entirely by the GM as an afterthought. There is a suggestion of no engagement with the fiction at all by the players! - which directly contradicts how the DMG says they are to be run.
(This ties into another issue with skill checks in 3E and 4e: I am becoming more and more persuaded that one function they are meant to serve is not as action resolution mechanics at all, but as techniques to allow the players to reframe the fictional situation rather than engage it. For example, a player who doesn't want to play through a social scene says instead "I roll Diplomacy" and if they succeed gets a new scene framed, in which there are no obstreporous NPCs to deal with. I'm a bit dubiuos about this sort of mechanic, but if it's going to be in the game I'd like it to be properly explained. Using
skill challenges as this sort of mechanic - which I gather is how at least some groups use them - strikes me as completely pointless, because it just drags out the process of reframing the scene.)