I want to share information about what I've been working on over the past few weeks. Most of these changes will interest developers who strive to use EyeAuras as a framework for creating their own products. This update will not affect users who do not use CustomUI/ExecuteScript.
Two months ago, I implemented a new type of overlay, CustomUI, which allows creating a user interface using C# and Razor (HTML/CSS). This has opened up a wide range of possibilities for experienced developers, giving them the ability to create a custom interface for their bot/set of auras. It simplifies the process of understanding and using the EyeAuras interface for a user who imports your set of auras.
An important limitation was that developers are used to working in development environments like Visual Studio and Rider, but previously, the ability to work with C# code was only available through the built-in EyeAuras editor. This editor has its limitations and cannot compete in quality and speed with more powerful IDEs. This is what I decided to change.
As soon as the alpha test ends, the new CustomUI overlay will have the ability to export code and save it as a C# project. The created project will include EyeAuras.SDK, which allows development using all the features available in the program: input simulation, image search, AI object detection, and much more. It's important to note - the code still needs EyeAuras running to launch.
This project can be opened in any IDE (Visual Studio, Rider, SharpDevelop, or even VS Code) and get full auto-completion, the ability to write unit tests, and all the other benefits of modern development. It's important to note that the ability to connect external Nuget packages will appear in the future. By default, EyeAuras already includes about 50 different libraries, which should be enough in most cases. However, I will definitely add this feature.
In addition to the ability to import the project into EyeAuras, there is also a Live development mode, in which EyeAuras tracks all the changes you are making in the project right now, picks them up, and loads them in real-time. I.e., from the moment you changed the style of the button, added some new logic to your bot or anything else, until the moment you saw them on your screen in the form of a working product, it takes on average 2-3 seconds. And given that this code has access to any triggers, current captured images with and without applied effects, all input simulators, etc., the possibilities are huge.
