Create

Sign in to ReadmeX

or

How we handle your data: Privacy Policy

ReadmeX

Your AI reading companion

What would you like to catch up on?

Pick a briefing, or ask your own question.

Your community and people briefings stay personal to you.

andrej-karpathy-skills

andrej-karpathy-skills: A CLAUDE.md file that improves Claude Code behavior by addressing common LLM coding pitfalls.

What problem it solves

Coding agents may silently make assumptions, fail to surface tradeoffs, overcomplicate implementations, and modify code they do not fully understand. The guidelines seek more focused changes, earlier clarification, simpler code, and verifiable execution.

Key features

  • Think Before Coding — The guidelines require explicit assumptions, clarification of ambiguity, consideration of multiple interpretations, and pushback when a simpler approach is available.

    Source: README

  • Simplicity First — The guidelines discourage speculative features, unnecessary abstractions, unrequested configurability, and error handling for impossible scenarios.

    Source: README

  • Surgical Changes — Changes should remain scoped to the request, preserve existing style, and avoid unrelated refactoring while removing only artifacts made obsolete by the change.

    Source: README

  • Goal-Driven Execution — Tasks are expressed as verifiable goals with tests or explicit checks, allowing implementation to proceed through a defined verification loop.

    Source: README

Quick start

Copied verbatim from the project's README.

/plugin marketplace add forrestchang/andrej-karpathy-skills
/plugin install andrej-karpathy-skills@karpathy-skills
curl -o CLAUDE.md https://raw.githubusercontent.com/forrestchang/andrej-karpathy-skills/main/CLAUDE.md

Who it may suit

This project may suit Claude Code users who want clearer assumptions, simpler implementations, and more focused edits. It may also suit teams seeking reusable guidance that can be merged with project-specific instructions.

Topics to discuss

  • Which guideline most effectively prevents unnecessary changes in coding-agent diffs?

  • How should teams balance the guidelines' caution with speed on trivial tasks?

  • Which project-specific rules work best alongside these general principles?


Written by ReadmeX AI from the project's README, source and public GitHub data; it may contain mistakes.

Read the original README https://github.com/multica-ai/andrej-karpathy-skills#readme

ReadmeX opened this community for a trending project; it is not run by the project. Maintainers can claim it from the community page.

github.comhttps://github.com/multica-ai/andrej-karpathy-skills

Comments

…