Basically what i want, is to build a voice command and control application(more like a virtual assistant), the voice recognition features and the natural language processing is already taken care of.

However i don't seem to find any specific way to control third party applications of a system(Windows, Linux or Mobile Devices).

simon seems to be doing similar stuff

From the stackoverflow threads i have found the following solutions to solve the problem

1.Autoit Scripting (Which is basically a automation tool for testing) https://www.autoitscript.com/site
2.Sikuli (it is similar software with advanced ocr and image processing) http://www.sikuli.org/
3.Vocola Language (it seemed perfect for my system however it seems to only support Dragon Naturally Speaking and Windows Speech Recognition) http://vocola.net/
4.Natlink (It is a python scripting module for dragon naturally speaking) https://sourceforge.net/projects/natlink/
5.Dragonfly (It is similar to vocola) https://github.com/t4ngo/dragonfly
6.User32.dll (which is a windows sdk library that can emulate user action)

so my question is how exactly simon executes a command in an application. what are the underlying technologies, frameworks and logic behind it?

thanks in advance.