A single CLAUDE.md file to transform your AI coding assistant, derived from Andrej Karpathy's observations on LLM coding pitfalls.
"The models make wrong assumptions on your behalf and just run along with them without checking. They don't manage their confusion, don't seek clarifications..."
"They really like to overcomplicate code and APIs, bloat abstractions, don't clean up dead code... implement a bloated construction over 1000 lines when 100 would do."
"They still sometimes change/remove comments and code they don't sufficiently understand as side effects, even if orthogonal to the task."
State assumptions explicitly, present multiple interpretations, push back when warranted, stop when confused.
Minimum code that solves the problem. No speculative features, no unnecessary abstractions, no overengineering.
Touch only what you must. Match existing style. Don't refactor things that aren't broken. Clean up only your own mess.
Define success criteria. Write tests first. Loop until verified. Transform "what to do" into "what success looks like".
Fewer unnecessary changes in diffs. Only requested changes appear in pull requests.
Fewer rewrites due to overcomplication. Code is simple and correct from the start.
Clarifying questions come before implementation, not after costly mistakes.
Clean, minimal PRs with no drive-by refactoring or unnecessary improvements.
Join thousands of developers already using these guidelines to build better, cleaner code with AI assistance.