karpathy-guidelines

Apply behavioral guidelines to LLM code development with verifiable success criteria.

3|Updated Mar 25, 2026

Search Tags:#code quality #code review #refactoring #code guidelines #LLM coding

Ready to run

npx skills add https://github.com/adihebbalae/Attacca --skill karpathy-guidelines-adihebbalae

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: karpathy-guidelines
Source: https://github.com/adihebbalae/Attacca/tree/main/addons/kiro/.kiro/skills/karpathy-guidelines
Command: npx skills add https://github.com/adihebbalae/Attacca --skill karpathy-guidelines-adihebbalae

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill provides behavioral guidelines to reduce common LLM coding mistakes, helping users avoid overcomplication, make surgical changes, surface assumptions, and define verifiable success criteria.

Core Features & Use Cases

Behavioral Guidelines: Offers a set of guidelines inspired by Andrej Karpathy to prevent overcomplication in LLM code.
Surgical Changes: Encourages editing only necessary parts of the code and keeping style consistency.
Goal-Driven Execution: Promotes defining clear success criteria and verifying goals for each task.

Quick Start

Apply the Karpathy guidelines when reviewing or refactoring code to improve its clarity and effectiveness.

Frequently Asked Questions about karpathy-guidelines

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

FAQPage Schema

How do I prevent common LLM coding mistakes during code generation?▼

To prevent common LLM coding mistakes, apply behavioral guidelines that enforce surgical changes, surface hidden assumptions, and define verifiable success criteria for each task. This approach reduces overcomplication and ensures code clarity.

What are the best guidelines for LLM code review and refactoring?▼

The best guidelines for LLM code review focus on making surgical changes, maintaining style consistency, and verifying that defined success criteria are met. This prevents overcomplication and ensures the generated code remains clear and effective.

How do I ensure LLMs make surgical changes to existing code?▼

You can ensure LLMs make surgical changes by applying behavioral guidelines that restrict edits to only necessary parts of the code. This approach maintains style consistency and prevents the model from overcomplicating unrelated logic during refactoring.

Why does LLM generated code become overcomplicated and how to fix it?▼

LLM generated code becomes overcomplicated when assumptions are not surfaced and success criteria are undefined. Fix it by applying behavioral guidelines that promote surgical changes, goal-driven execution, and explicit verification for each coding task.

Do I need specific tools to apply Karpathy guidelines for code quality?▼

You do not need specific tools to apply Karpathy guidelines for code quality. The guidelines are behavioral rules for LLM code development that require no dependencies, focusing on making surgical changes and verifying success criteria directly.