You can paste some Korean text into https://community.languagetool.org/analysis/index?lang=en, and see what happens: the analysis will be wrong, but as words are separated by spaces in Korean, it should recognize the words. So you can indeed use the English
grammar.xml to add rules for Korean (after deleting all the English rules). In a first step, you can use words and regular expressions, you cannot yet refer to the words' part-of-speech. Still, I'd suggest to start like that, as it requires no setup at all, just download LT and edit
Once that works, we can think about how to activate part-of-speech analysis. We need either a dictionary for that or an existing Open Source library that can do this already.