Session Extractions
Six questions that remove work before you write a line of code — and a CLAUDE.md block that teaches the AI to propose less, not more.
Teaching Blocks
A meta-analysis of the reasoning pattern behind a clean refactor — six questions that each remove work before you write a line of code, formalized into a thinking skill you can run on your own architecture decisions.
Read →How a single back-and-forth about over-engineering turned into a paste-ready CLAUDE.md block that filters what the assistant proposes — calibrating suggestion volume to phase and stakes rather than gagging the model entirely.
Read →