karpathy-guidelines

Apply behavioral guidelines for minimal, verifiable coding changes.

Updated Mar 3, 2026
One-click install
npx skills add https://github.com/DonYum/yf-skills --skill karpathy-guidelines-donyum
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: karpathy-guidelines
Source: https://github.com/DonYum/yf-skills/tree/main/skills/karpathy-guidelines
Command: npx skills add https://github.com/DonYum/yf-skills --skill karpathy-guidelines-donyum

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Behavioral guidelines to reduce common LLM coding mistakes during writing, reviewing, or refactoring, promoting clearer reasoning and safer edits.

Core Features & Use Cases

  • Think Before Coding: surface assumptions, ask questions, and surface tradeoffs.
  • Simplicity First: deliver minimal, well-scoped solutions without unnecessary abstractions.
  • Surgical Changes: modify only the affected parts; avoid unintended side effects and preserve existing structure.
  • Goal-Driven Execution: define verifiable success criteria and loop until they are met, with explicit plans and checks.

Quick Start

Apply the Karpathy guidelines to your current coding task to ensure minimal, verifiable changes.

Frequently Asked Questions about karpathy-guidelines

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

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

To reduce LLM coding mistakes, apply behavioral guidelines that enforce surgical changes, modifying only affected code parts while preserving existing structure to avoid unintended side effects. This ensures minimal, well-scoped edits for predictable refactoring outcomes.

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

Effective guidelines for LLM code review prioritize thinking before coding by surfacing assumptions, preferring simplicity without unnecessary abstractions, and pursuing goal-driven execution with verifiable success criteria. This approach enforces explicit assumptions and minimal edits.

How do I make surgical changes to code using an LLM?

Make surgical changes by instructing the LLM to modify only the affected parts of the code while avoiding unintended side effects. This approach preserves existing structure and ensures minimal, well-scoped edits for predictable outcomes.

How do I enforce goal-driven execution when generating code with an LLM?

Enforce goal-driven execution by defining verifiable success criteria and looping until they are met. This requires the LLM to establish explicit plans and checks, ensuring the generated code fulfills the intended objectives predictably.

Why does my LLM generate overly complex code with unnecessary abstractions?

LLMs generate overly complex code when not constrained by simplicity-first guidelines. To prevent this, enforce rules that require the LLM to deliver minimal, well-scoped solutions and surface tradeoffs before writing code.