andrej-karpathy-skills

Community

Make changes small, verifiable, and safe.

Authordmae97
Version1.0.0
Installs0

System Documentation

What problem does it solve?

It solves the problem of unreliable coding outcomes by enforcing a disciplined, goal-driven workflow that minimizes regressions and proves success with observable checks.

Core Features & Use Cases

  • Assumption clarification: surfaces only the assumptions that materially affect implementation, asking for clarification when ambiguity blocks safe progress.
  • Evidence-based success criteria: converts “do the change” requests into verifiable outcomes such as tests, lint, typecheck, build, screenshots, or exact output.
  • Smallest-scope execution: limits file touchpoints and avoids unrelated cleanup so refactors, bugfixes, and debugging remain tightly bounded.

Quick Start

Use this skill when you want an AI to implement a refactor or bugfix with a minimal diff and clear success checks for the changed behavior.

Dependency Matrix

Required Modules

None required

Components

Standard package

💻 Claude Code Installation

Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.

Please help me install this Skill:
Name: andrej-karpathy-skills
Download link: https://github.com/dmae97/oh-my-kimi/archive/main.zip#andrej-karpathy-skills

Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
View Source Repository

Agent Skills Search Helper

Install a tiny helper to your Agent, search and equip skill from 471,000+ vetted skills library on demand.