multica-ai/andrej-karpathy-skills
Repository · https://github.com/multica-ai/andrej-karpathy-skills
A single CLAUDE.md file to improve Claude Code behavior, derived from Andrej Karpathy's observations on LLM coding pitfalls.
First seen by Oriel on .
AI summary
This repository packages four Karpathy-inspired principles in a CLAUDE.md file and Claude Code plugin to reduce mistaken assumptions, overengineering, unrelated edits, and unverified work.
- Think Before Coding asks Claude to state assumptions, surface ambiguity and tradeoffs, and request clarification when needed.
- Simplicity First discourages unrequested features, single-use abstractions, speculative flexibility, and excessive error handling.
- Surgical Changes limits edits to the task and cleanup caused by those edits, while preserving unrelated code and comments.
- Goal-Driven Execution turns tasks into verifiable success criteria, such as writing a reproducing test before fixing a bug.
- The guidelines can be installed as a Claude Code plugin or added to a project’s CLAUDE.md; a committed Cursor rule is also included.
#claude code #llm agents #software engineering
Generated by gpt-5.6-sol. Check the original for details.
Metric trends
Last 30 days · Up to 120 recorded samples per source and metric, with reposts kept separate · First and latest retained
Stars
Total repository stars at each observation.
- GitHub Trending: 225,099, latest observation ; +978 over observed period since
Stars gained (1d)
Daily star gain reported by GitHub Trending. A change compares two daily values; it is not additional stars between observations.
- GitHub Trending: 638, latest observation ; +181 over observed period since
Forks
Total repository forks at each observation.
- GitHub Trending: 22,502, latest observation ; +73 over observed period since
Aliases
| Source | External ID | URL |
|---|---|---|
| GitHub Trending | multica-ai/andrej-karpathy-skills | https://github.com/multica-ai/andrej-karpathy-skills Read discussion |