Incase anyone is still using Eule.py: In November 2020 I did a (nearly, the abbrevation feature is missing) complete implementation of it using WPF in C#.
It can be found here:
GitHub - VocalTrance/DolphinD3.
The programm is smaller, much more performant and the implementation is now atleast following some patterns (MVVM, Microservices, Publish-Subscribe, Dependency Injection).
Besides that, it contains some condition based Macros for skill casting, which is fully image recognition driven (this, together with learning some C#, initially was my goal).
However, using only image recognition proved to be too unrealiable and too big of an effort for different resolutions.
The information exploitation (checking conditions) is stricly decoupled from the information generation (image recognition) so if anyone comes around with a better method for generating the information needed, it would fit rirght in.
If you still want to try the skill casting it in its current form:
Download the project, checkout dev branch and compile the Dolphin.Ui project.
It is currently implemented for the skills listed in 2560x1440p resolution in greater rifts.
You can check if the image recognition is working for you by checking the Overview Tab (if the boxes are empty, it isnt working ;D).
Im posting this because I saw some new comments and thought that people might still be interested in this kind of QOL 3rd party diablo software.
I'm not currently working on the project anymore, but I saw some people make changes in their forks of Eule.py and thought that you may also enjoy (or enjoy it more) to make those adjustments on the Dolphin project instead.
I dont really have time to work on the project myself right now, but might pick it up again if people are liking it.