Andrej Karpathy Skills

Plugin Verified Active

Part of:Karpathy Inspired Claude Code Guidelines

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

1 Skill 0 MCPs

Purpose

To help users improve LLM code generation by providing clear principles to avoid common mistakes like overcomplication, incorrect assumptions, and surgical errors.

Features

Provides four core principles for better LLM coding: Think Before Coding, Simplicity First, Surgical Changes, and Goal-Driven Execution.
Offers detailed explanations and actionable advice for each principle.
Can be used as a Claude Code plugin or directly via a CLAUDE.md file.
Includes guidance on defining success criteria for LLM tasks.

Use Cases

When generating new code with an LLM to ensure it follows best practices from the start.
When reviewing code produced by an LLM to identify potential pitfalls.
When refactoring existing code to maintain simplicity and avoid unnecessary changes.
When working on projects where consistent, high-quality code output from LLMs is crucial.

Non-Goals

Providing language-specific linting or code formatting rules.
Automating code execution or deployment processes.
Acting as a general-purpose LLM prompt engineering tool beyond coding practices.

Installation

First, add the marketplace

/plugin marketplace add forrestchang/andrej-karpathy-skills

/plugin install andrej-karpathy-skills@karpathy-skills

Contains 1 extensions

Skill (1)

Karpathy Guidelines Skill

Behavioral guidelines to reduce common LLM coding mistakes. Use when writing, reviewing, or refactoring code to avoid overcomplication, make surgical changes, surface assumptions, and define verifiable success criteria.

Quality Score

Verified

99 /100

Analyzed 6 days ago

Trust Signals

Last commit29 days ago

GitHub owner forrestchang

Stars128.1k

LicenseMIT

Status

View Source

Similar Extensions

Context7 Plugin

100

Upstash Context7 MCP server for up-to-date documentation lookup. Pull version-specific documentation and code examples directly from source repositories into your LLM context.

Plugin

upstash

LLM Wiki

Turn Claude Code + Obsidian into a second brain. The LLM incrementally ingests sources into a persistent, interlinked markdown wiki — building entity/concept/source pages, flagging contradictions, maintaining an index and log. Knowledge compounds instead of being re-derived by RAG on every query. Inspired by Karpathy's LLM Wiki gist. Ships SKILL, 3 sub-agents, 5 slash commands, 8 Python tools (stdlib only), full vault templates, and cross-tool compatibility (Claude Code, Codex CLI, Cursor, Antigravity, OpenCode, Gemini CLI).

Plugin

alirezarezvani

Framework Migration

Framework updates, migration planning, and architectural transformation workflows

Plugin

wshobson

Dimensional Analysis

Annotates codebases with dimensional analysis comments documenting units, dimensions, and decimal scaling. Use when someone asks to annotate units in a codebase, perform a dimensional analysis, or find vulnerabilities in a DeFi protocol. Prevents dimensional mismatches and catches formula bugs early.

Plugin

trailofbits

Codebase Cleanup

Technical debt reduction, dependency updates, and code refactoring automation

Plugin

wshobson

Brand Voice

Brand Voice transforms scattered brand materials into enforceable AI guardrails — automatically. It searches across Notion, Google Drive, Confluence, Gong, Slack, and meeting transcripts to distill your strongest brand signals into a single source of truth, then applies them to every piece of AI-generated content. The more your team creates with Claude, the more consistent your brand becomes.

Plugin

anthropics