I just searched for a way of allowing pagebreaks when using shaded theorem-environments from thmtools. I found my question was, somewhat, answered here (the second solution) with a comment from the package's author Ulrich Schwarz. However, I'd like to keep defining theorem-like environments as follows
\declaretheorem[shaded={bgcolor=LightGrey},name=Definition,parent=chapter,
% refname={definition,definitions},
Refname={Definition,Definitions}]{Def}
i.e. without having to declare an extra (rather bulky) theoremstyle for each environment I define.
Is it possible to use the "preheadhook-postfoothook-trick" with mdframed globally, while retaining the overall interface and style I would get with my above \declaretheorem-example?
As I understand, shaded theorem-environments in thmtools are already implemented using mdframed anyway, so I thought it should be possible to allow pagebreaks by redefining some internal thmtools command, by setting some global option, or something along those lines.
I hope this is actually a valid question, as it was, more or less, already answered in 2011.

mdframedmay have some issues with theorem break that ends up looking odd (such as placing the box end on the next page, not sure if this is an issue when combined withthmtools). So for my own projects I now use unbreakble theorems and also provide theorems that are breakable and then manually change the environment. That seems to provide better results. – daleif Apr 16 '15 at 11:36