Evaluation Framework for Agentic Coding Tools and Practices

This framework helps practitioners evaluate AI coding tools and development practices. It is designed to remain useful as the landscape changes -- which it does every 3 months.

How to Use This Framework

Score each dimension 1-5 for any tool or practice you are evaluating. The total score is less important than the profile: a tool that scores 5 on autonomy and 1 on methodology support is a different kind of tool than one that scores 3 on both.