karpathy-guidelines

Apply behavioral guidelines to reduce LLM coding mistakes during development, review, and refactoring.

Updated Apr 11, 2026
One-click install
npx skills add https://github.com/zillafan80-Maxzilla/claw-code-ubuntu-desktopUI --skill karpathy-guidelines-zillafan80-maxzilla
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: karpathy-guidelines
Source: https://github.com/zillafan80-Maxzilla/claw-code-ubuntu-desktopUI/tree/main/payload/claw-code/.claw/skills/karpathy-guidelines
Command: npx skills add https://github.com/zillafan80-Maxzilla/claw-code-ubuntu-desktopUI --skill karpathy-guidelines-zillafan80-maxzilla

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Behavioral guidelines to reduce common LLM coding mistakes, derived from Andrej Karpathy's observations on LLM coding pitfalls.

Core Features & Use Cases

  • Clear rules to avoid overcomplication and ensure surgical changes.
  • Surface assumptions, define verifiable success criteria.
  • Applicable across writing, reviewing, and refactoring tasks in varied project contexts.

Quick Start

Apply these guidelines to a coding task by stating assumptions, keeping changes minimal, and defining explicit success criteria.

Frequently Asked Questions about karpathy-guidelines

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I reduce common LLM coding mistakes during refactoring?

To reduce LLM coding mistakes during refactoring, apply behavioral guidelines that surface assumptions, minimize changes to ensure surgical edits, and define verifiable success criteria for the codebase.

What are the best guidelines for LLM code review and generation?

Effective guidelines for LLM code review involve avoiding overcomplication, enforcing surgical changes, stating explicit assumptions, and defining measurable success criteria to verify code behavior across varied project scopes.

Why does my LLM overcomplicate code changes and how do I fix it?

LLMs overcomplicate code changes when lacking strict boundaries; fix this by applying surgical change guidelines that minimize edits, surface underlying assumptions, and define verifiable success criteria before execution.

Can I apply these LLM coding guidelines to any project scope?

Yes, you can apply these LLM coding guidelines across writing, reviewing, and refactoring tasks in varied project contexts by consistently surfacing assumptions and defining explicit success criteria.

How do I define verifiable success criteria for LLM coding tasks?

Define verifiable success criteria for LLM coding tasks by explicitly stating assumptions and establishing clear, measurable rules that validate the surgical changes made during development or refactoring.