I just use Opus with the simple-english skill. No more Claudese.
Nah, I haven't found complete success with tweaks like that. I would strongly suspect they're still there and you might just not be as sensitive to them.
I would bet you still get tells like overusing "never" or ubiquitous inanimate subjects with animate verbs ("questions that decide the option").
I've been maintaining SlopSift - https://slopsift.dev/ for a while it's powered by a dependency parser to deterministically lint English. Then I wrap it around Claude either as a skill or a hook. It's really simplified the AI speak for me
loading story #50557062
loading story #50561048
Simple English skills get rid of the most egregious word choices but their awkward sentence structure still comes through for me. I can definitely still tell, if only because I know what my coworker's actual writing sounds like.
loading story #50559682
Could you link it, if it's open? I haven't had good success with ASD-STE100 myself :(
loading story #50557736
This is better, but you might try the Gemini models if you have access. They are much better writers, even with the simple English skill.