karpathy-guidelines

Apply behavioral guidelines for writing, reviewing, and refactoring code.

Updated Jun 6, 2026
One-click install
npx skills add https://github.com/kenhia/kdeskdash --skill karpathy-guidelines-kenhia
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: karpathy-guidelines
Source: https://github.com/kenhia/kdeskdash/tree/main/.github/skills/karpathy-guidelines
Command: npx skills add https://github.com/kenhia/kdeskdash --skill karpathy-guidelines-kenhia

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill offers behavioral guidelines to avoid common LLM coding mistakes, promoting simplicity, clarity, and maintainability in your codebase.

Core Features & Use Cases

  • Guidelines for LLM Coding: Reduces overcomplication, ensures clarity, and sets verifiable success criteria.
  • Code Review and Refactoring: Enhances code quality during reviews and refactoring.
  • Use Case: When you're writing, reviewing, or refactoring code to avoid pitfalls often encountered with LLMs.

Quick Start

Invoke the karpathy-guidelines skill to apply behavioral guidelines to your coding process.

Frequently Asked Questions about karpathy-guidelines

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

FAQPage Schema
What are common LLM coding mistakes when writing or refactoring code?

Common LLM coding mistakes include overcomplicating logic, lacking clarity, and missing verifiable success criteria. Behavioral guidelines help avoid these pitfalls by promoting simplicity and maintainability during code generation and refactoring.

How do I apply programming best practices for LLM code review?

To apply programming best practices for LLM code review, invoke behavioral guidelines that enforce simplicity, clarity, and verifiable success criteria. This ensures the generated code meets maintainability standards before merging.

Can I use code guidelines to simplify refactoring tasks?

Yes, you can use code guidelines to simplify refactoring tasks. They provide behavioral rules for LLM coding that reduce overcomplication and ensure the refactored code remains clear, simple, and maintainable.

What is the best way to set verifiable success criteria for LLM generated code?

The best way to set verifiable success criteria for LLM generated code is to apply behavioral guidelines that enforce clear, testable outcomes. This prevents overcomplication and ensures the code meets functional requirements.

Does this approach work for general software engineering or only LLM coding?

This guidelines approach works for general software engineering but is specifically designed for LLM coding. It targets common pitfalls of LLM-generated code, ensuring simplicity and clarity during writing and refactoring.