can the function of obsidian-smart-connections be implemented outside of obsidian?

[PaxPax40]

thanks for this wonderful software.

(not a feature/solution request, just a general query)

asking whether such a solution exists already or if I am not looking harder since I am not tech-oriented, I want there to be a script that on being invoked, takes a partial screenshot of my screen, performs OCR on it and makes an embedding(?) of the ocr result which is progressively added to each time I run the script.

The script can prompt me the option to choose from whether to:

1. embed the contents of the screenshot, or...
2. to embed and search for semantically similar notes, or...
3. to embed, search AND ask a question based on the semantically similar notes as context.

currently, I have a very crude implementation of this which involves scripts for taking screenshot and saving the OCR result as md files inside an obsidian vault where I can perform all the required functions, I am barely inconvenienced by it since it gets the job done very well, but at the same time prompting gemini to create me these neat little scripts which help with my workload tremendously, makes me believe that it should be possible to implement something like this given I know the logic of it.