This is a structured framework for validating startup ideas before you write code. It walks you through defining testable hypotheses, collecting evidence, scoring opportunities across dimensions like market size and willingness to pay, and arriving at a clear GO/CONDITIONAL/PIVOT/NO-GO decision. The 2026 operating principles are refreshingly opinionated: separate evidence quality from confidence, pre-register your thresholds before experiments to avoid moving goalposts, and calibrate everything to your actual target outcome. It's designed for founders who want to kill bad ideas fast or build conviction around good ones with something more rigorous than gut feel. Over 200 installs suggests people are hungry for this kind of systematic approach.
npx skills add https://github.com/vasilyu1983/ai-agents-public --skill startup-idea-validation