Scientific Problem Selection

This is a conversational framework for scientists to think through research problem selection, based on Fischbach & Walsh's work in Cell. It gives you three entry points: pitch a new idea, troubleshoot a stuck project, or ask a strategic question. From there it walks you through evaluating feasibility, defining what success looks like, deciding which parameters to fix versus leave flexible, and planning decision points. The core insight is that scientists typically spend days choosing problems and years solving them, which is backwards given that problem choice matters more than execution quality. It's designed to help you invest more upfront time in choosing wisely, with modules for risk assessment, adversity planning, and navigating around obstacles when things inevitably go sideways.

Install to Claude Code

npx -y skills add anthropics/knowledge-work-plugins --skill scientific-problem-selection --agent claude-code