I can't guess why Magic Sword thinks 5e is easy to create ad hoc rulings for heists, but I can absolutely say it's because there's nothing there to interfere with doing so because 5e has no structure at all to bump into. I was trying to not put words in MS's mouth, not weaseling on the point that 5e lacks any support for heist play.
Is that sufficiently adamantium-mithril allow?
So I'm not sure how much the conversation moved on after this page, but to clarify since I think this thing about our posts deserves a proper answer from me, I think the subtle difference between our posts people were latching onto is your suggestion that the modification is possible because of an
absence of rules, which isn't quite true.
In DND, there are canonical answers to those rules questions in the post you make below-- if you want to sneak past someone, you roll a stealth check; if you want to bribe the guard, you probably roll a diplomacy check. The difference between the two rulesets, and what makes 5e easier to 'adhoc' is that DND presents and structures its rules options as a toolkit instead of as an explicit procedure. BITD (which I've recently bought and been reading, it seems really fun) expects you to use its role procedure and has more elements of that hang off each other.
If you're playing BITD and the GM starts editing the procedure for how heists work for a different experience, the game itself is a little more hostile to that, because how it works is interrelated in a mechanical context-- if you don't use the flashbacks in favor of a 'full planning session' approach, it affects how stress plays out, which ties into the games other systems. If you don't use the engagement roll, then you're now coping with how that affects the footing system as well.
DND gives you solutions and tools to run heist stuff, like when I play Pathfinder 2e (which is my 'DND system' of choice) I could very reasonably just use
the infiltration subsystem or if I'm playing 4e, it could just be a Skill Challenge (X Successes before Y Failures) (both of which are structured to guide the DM away from calling for too many checks), or I could run it off the cuff, as a series of stealth and deception checks (which is a challenge I would have to curate accordingly to keep the odds of success decent, I'd lean toward letting each category of obstacle be a check, rather than roll separately for each guard) which would be 5e's canonical answer.
The difference is that nothing says I have to run it any one of these ways, the system encourages me to pick one if it suits me, or make another, and it includes optional 'alternatives' if I want a more refined take sometimes, the Saltmarsh sailing rules are different than the default ones, for instance. Personally, I think 5e has a flaw in that it tries to be too rules lite, so it doesn't like to provide subsystems for stuff like this, but these other systems are DND games as well, and they demonstrate the reality of it. I'm not just benefitting from a lack of rules, I'm benefitting from an intentional choice of rules, the system goes out of its way to make itself resilient to modification for stuff like this.