Karpathy Inspired Claude Code Guidelines
Marketplace Verified ActiveBehavioral guidelines to reduce common LLM coding mistakes, derived from Andrej Karpathy's observations
To help users improve the quality and focus of AI-generated code by applying established principles that prevent common LLM coding errors.
Features
- Actionable coding guidelines derived from expert observations
- Addresses common LLM pitfalls: incorrect assumptions, overcomplication, orthogonal edits
- Provides four core principles: Think Before Coding, Simplicity First, Surgical Changes, Goal-Driven Execution
- Offers integration via Claude Code plugin or CLAUDE.md file
Use Cases
- Use when AI-generated code frequently contains errors or unexpected changes.
- Incorporate into projects to enforce consistent coding quality from AI assistants.
- Leverage to guide AI assistants toward more focused and verifiable task execution.
Non-Goals
- Providing specific code implementations for tasks.
- Acting as a general-purpose code formatter or linter.
- Replacing the need for human code review entirely.
Practices
- Coding quality
- LLM behavior
- Prompt engineering
Installation
/plugin marketplace add forrestchang/andrej-karpathy-skillsQuality Score
VerifiedTrust Signals
Similar Extensions
Caveman Repo
100🪨 why use many token when few token do trick — Claude Code skill that cuts 65% of tokens by talking like caveman
Cozempic
1002x the effective context with smart weight-loss for Claude Code — prune bloated sessions, protect agent teams from compaction, monitor token usage with MCP tools
Cortex Plugins
100Persistent memory and cognitive profiling plugins for Claude Code
Stripe
100One-stop shop for building AI-powered products and businesses with Stripe.
Context Mode
100Context window optimization for AI coding agents. Sandboxes tool output, 98% reduction. 15 platforms