Claude Agent Skill · by Forrestchang

Karpathy Guidelines

The karpathy-guidelines skill provides behavioral guidelines for LLM code generation that reduce common mistakes by emphasizing explicit assumptions, simplicity

Install
Terminal · npx
$npx skills add https://github.com/forrestchang/andrej-karpathy-skills --skill karpathy-guidelines
Works with Paperclip

How Karpathy Guidelines fits into a Paperclip company.

Karpathy Guidelines drops into any Paperclip agent that handles this kind of work. Assign it to a specialist inside a pre-configured PaperclipOrg company and the skill becomes available on every heartbeat — no prompt engineering, no tool wiring.

S
SaaS FactoryPaired

Pre-configured AI company — 18 agents, 18 skills, one-time purchase.

$27$59
Explore pack
Source file
SKILL.md67 lines
Expand
---name: karpathy-guidelinesdescription: 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.license: MIT--- # Karpathy Guidelines Behavioral guidelines to reduce common LLM coding mistakes, derived from [Andrej Karpathy's observations](https://x.com/karpathy/status/2015883857489522876) 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.** When editing existing code:- Don't "improve" adjacent code, comments, or formatting.- Don't refactor things that aren't broken.- Match existing style, even if you'd do it differently.- If you notice unrelated dead code, mention it - don't delete it. When your changes create orphans:- Remove imports/variables/functions that YOUR changes made unused.- Don't remove pre-existing dead code unless asked. The test: Every changed line should trace directly to the user's request. ## 4. Goal-Driven Execution **Define success criteria. Loop until verified.** Transform tasks into verifiable goals:- "Add validation" → "Write tests for invalid inputs, then make them pass"- "Fix the bug" → "Write a test that reproduces it, then make it pass"- "Refactor X" → "Ensure tests pass before and after" For multi-step tasks, state a brief plan:```1. [Step] → verify: [check]2. [Step] → verify: [check]3. [Step] → verify: [check]``` Strong success criteria let you loop independently. Weak criteria ("make it work") require constant clarification.