What is kind?

Kind is a very small, raw proof language built on top of Type Theory. Type Theory is a branch of mathematics and computer science concerned with formally proving that programs are correct. Kind strips that idea down to its most basic form, giving you a minimal foundation rather than a full-featured system. The project is written in Haskell and is used via a command-line tool called kind. To get started, you clone the repository, install it, and then run the kind command to check or evaluate terms you write. The README for this project is mostly a formal grammar specification. It describes the building blocks of the language: types of terms such as functions, applications, type annotations, self types, data types, constructors, pattern matches, numeric operations, and holes. If you have worked with a functional or proof language before, this notation will look familiar. If you have not, it reads as a compact rule-set that defines every valid piece of syntax the language accepts. The documentation provided is minimal and technical. There is no tutorial, no worked examples, and no explanation of what problems the language is intended to solve beyond "a very raw and minimal Type Theory." This is clearly a project aimed at researchers or developers already comfortable with formal type systems who want a small, auditable core they can reason about or build upon. If you are looking for a beginner-friendly proof assistant, this is not the right starting point. If you want a stripped-down Type Theory core to study or extend, Kind offers exactly that with very little overhead on top of the fundamental ideas.