I created a topic on the old forums, archived here, regarding lookalike kanji giving me grief one day. Someone in the thread kindly provided a reddit link. While I take a short break from compiling the stuff for the corpus script, I whipped this up. It more or less works already so I’m wondering if there is any interest in going forward with it. I’m using the reddit list right now with two entries of my own (‘仕休’ & ‘主玉王’), and am certainly willing to add more.
If you’re as bad at paying attention as I am sometimes – it’s the “Not: …” bit that this script is adding.