Based on Andrej Karpathy's observations about common LLM coding mistakes, this skill pushes back against the tendency to overcomplicate. It enforces four core behaviors: surface your assumptions and confusion before writing code, bias toward minimal solutions without speculative features, make surgical edits that touch only what's necessary, and define verifiable success criteria for every task. The tradeoff is explicit: you sacrifice speed for caution. For simple scripts this might feel heavy, but for production code or refactors where LLMs typically add unnecessary abstractions or touch too many files, the discipline pays off. Think of it as guardrails against your AI pair programmer's worst habits.

install copies · 90 days. Times the install command was copied on this site in the last 90 days, bots filtered.
42
skills.sh installs
45.6K
GitHub stars
231.5K
last updated
Aug 21, 2026

Install

$ npx -y skills add multica-ai/andrej-karpathy-skills --skill karpathy-guidelines --agent claude-code

Installs into .claude/skills of the current project.

Files

Karpathy Guidelines

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

Tradeoff: These guidelines bias toward caution over speed. For trivial tasks, use judgment.

1. Think Before Coding

Don't assume. Don't hide confusion. Surface tradeoffs.

Before implementing:

  • State your assumptions explicitly. If uncertain, ask.
  • If multiple interpretations exist, present them - don't pick silently.
  • If a simpler approach exists, say so. Push back when warranted.
  • If something is unclear, stop. Name what's confusing. Ask.

2. Simplicity First

Minimum code that solves the problem. Nothing speculative.

  • No features beyond what was asked.
  • No abstractions for single-use code.
  • No "flexibility" or "configurability" that wasn't requested.
  • No error handling for impossible scenarios.
  • If you write 200 lines and it could be 50, rewrite it.

Ask yourself: "Would a senior engineer say this is overcomplicated?" If yes, simplify.

3. Surgical Changes

Touch only what you must. Clean up only your own mess.

Sponsors

Details

Repository
github.com/multica-ai/andrej-karpathy-skills
Categories
Documentation, Code Review & Quality, AI & Agent Building
First seen
May 16, 2026
1samber avatargolang-testingsamber/cc-skills-golangProduction-ready Golang tests — table-driven tests, testify suites and mocks, parallel tests, fuzzing, fixtures, goroutine leak detection with goleak, snapshot 45.4Kskills.sh installs
2samber avatargolang-securitysamber/cc-skills-golangSecurity best practices and vulnerability prevention for Golang. Covers injection (SQL, command, XSS), cryptography, filesystem safety, network security, cookie44.2Kskills.sh installs
3samber avatargolang-concurrencysamber/cc-skills-golangGolang concurrency patterns. Use when writing or reviewing concurrent Go code involving goroutines, channels, select, locks, sync primitives, errgroup, singlefl43.6Kskills.sh installs
4samber avatargolang-namingsamber/cc-skills-golangGo (Golang) naming conventions — covers packages, constructors, structs, interfaces, constants, enums, errors, booleans, receivers, getters/setters, functional 42.5Kskills.sh installs
5samber avatargolang-project-layoutsamber/cc-skills-golangProvides a guide for setting up Golang project layouts and workspaces. Use when starting a new Go project, organizing an existing codebase, setting up a monorep42.1Kskills.sh installs
6samber avatargolang-structs-interfacessamber/cc-skills-golangGolang struct and interface design patterns — composition, embedding, type assertions, type switches, interface segregation, dependency injection via interfaces41.3Kskills.sh installs