>On a more positive note: is there something that detects or warns against hostile tone to help write in a friendly and non-hostile way? :)

Similar idea for public figures: ML that can detect phrases that can be taken out of context easily.