帖子
Someone open-sourced an ADHD agent skill for Claude Code that rewires how agents handle fuzzy debugging, naming, and API design. The issue with standard autoregressive reasoning is that Chain-of-Thought anchors hard,once an LLM picks a direction, it sticks.
1mo
7 赞0 踩1 转发2 评论
?
评论
还没有评论。来抢沙发吧！