coding

# Cli taste of AA
- Use pnpm as the package manager for CLI projects. Confidence: 1.00
- Use TypeScript for CLI projects. Confidence: 0.95
- Use tsup as the build tool for CLI projects. Confidence: 0.95
- Use vitest for testing CLI projects. Confidence: 0.95
- Use Commander.js for CLI command handling. Confidence: 0.95
- Use clack for interactive user input in CLI projects. Confidence: 0.95
- Check for existing CLI name conflicts before running npm link. Confidence: 0.95
- Organize CLI commands in a dedicated commands folder with each module separated. Confidence: 0.95
- Include a small 150px ASCII art welcome banner displaying the CLI name. Confidence: 0.95
- Use lowercase flags for version and help commands (-v, --version, -h, --help). Confidence: 0.85
- Start projects with version 0.0.1 instead of 1.0.0. Confidence: 0.85
- Version command should output only the version number with no ASCII art, banner, or additional information. Confidence: 0.90
- Read CLI version from package.json instead of hardcoding it in the source code. Confidence: 0.75
- Always use ora for loading spinners in CLI projects. Confidence: 0.95
- Use picocolors for terminal string coloring in CLI projects. Confidence: 0.90
- Use Ink for building interactive CLI UIs in CommandCode projects. Confidence: 0.80
- Use ink-spinner for loading animations in Ink-based CLIs. Confidence: 0.70
- Hide internal flags from help: .addOption(new Option('--local').hideHelp()). Confidence: 0.90
- Use pnpm.onlyBuiltDependencies in package.json to pre-approve native binary builds. Confidence: 0.60
- Use ANSI Shadow font for ASCII art at large terminal widths and ANSI Compact for small widths. Confidence: 0.85
- Use minimal white, gray, and black colors for ASCII art banners. Confidence: 0.85
- Check if package is publishable using `npx can-i-publish` before building or publishing. Confidence: 0.85

---
name: karpathy-guidelines
description: 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:
\
Strong success criteria let you loop independently. Weak criteria ("make it work") require constant clarification.