This is a structured approach to gathering input from multiple sources, synthesizing it into a coherent plan, and validating it back with the people who'll actually use or execute it. You'd call it when stakeholder feedback is scattered, when you're about to ship something without enough user input, or when a solution needs buy-in from multiple parties. It walks you through collect, systematize, return for feedback, test in practice, then iterate. The implementation is opinionated about avoiding two failure modes: ignoring users entirely (command-ism) or just doing whatever the loudest voice says (tail-ism). It also maps the concept to AI work by treating code, logs, and git history as information sources worth consulting alongside human input. Useful when you feel like you're building in a vacuum.
npx skills add https://github.com/hughyau/qiushi-skill --skill mass-line