Like I said, I use both the ‘Regex Filter’ and ‘Regex Replacer’ functions, both of which do indeed already come with Textractor.
For the ‘Regex Filter’ I merely use two of the filters from the github FAQ (\s and <.+?>), while with ‘Regex Replacer’ I use this: |REGEX|(^\w+)[「『(]|BECOMES|【$1】|MODIFIER||END| which helps to isolate character names (as in, inside name boxes) from actually spoken text. I also use the ‘Replacer’ function simply to replace said names before they go to translation (and are typically garbled).