[Bug] Claude Code reasoning quality degradation in versions 2.1.8+

Bug Description
claude seems to have become "dumber" in it's choices lately. maybe past week or so? this is down from two weeks ago where it seemed like claude had a major uptick in smartness (corresponding to making ultrathink the default). seems 2.1.7 was really good and 2.1.13 and later is not that great. I feel like it keeps missing things and I have to guide it more. around 2.1.7 it was getting to be like, "this is really cool, it's guessing everything right 95% of the time". Now it's even below the 85% that I've gotten used to. Maybe I would say it's down to 75% accuracy... :(

Anyone else feel this recently?

Environment Info

• Platform: win32
• Terminal: windows-terminal
• Version: 2.1.22
• Feedback ID: ab099d5b-d766-4c76-94d8-b3c2794028e1