From the opposite point of view, before writing a significant patch/feature, it may be worth opening an issue with a quick description of what you'd like to imp

From the opposite point of view, before writing a significant patch/feature, it may be worth opening an issue with a quick description of what you'd like to implement to see if there's interest in it from the maintainers before investing the time in writing it. It may even be something someone is already working on in a private branch. Even if it's something you really need and plan on maintaining your own fork if rejected, you could still get some tips on how to best implement it or potential tricky bits to be aware of.