Last commit on April 14, 2026·Created on April 14, 2026
vtroisWhite/andrej-karpathy-skills
“Behavioral constraints to reduce common LLM coding mistakes and overcomplication.”
Combined rank
#539
across all skills
Stars
414
7d change unavailable
Forks
72
Watchers
1
Traction scoreGitHub stars can be faked, so popularity alone can be misleading. Traction Score looks for broader signs of real attention, adoption, and active maintenance.
TL;DR
These guidelines provide a framework for interacting with LLMs during the writing, reviewing, and refactoring of code. By prioritizing explicit assumptions over silent interpretations, the skill helps developers avoid speculative features and unnecessary abstractions.
The approach biases toward caution and simplicity, requiring the model to surface tradeoffs and define verifiable success criteria before implementation begins.
WHO IT'S FOR
AI agent / automation builders
optimizing LLM system prompts for coding
Technical leads / architects
establishing AI coding standards for teams
Software engineers using AI assistants
reviewing and refactoring LLM-generated code
QA / test automation engineers
defining verifiable success criteria for AI tasks
Compatible AgentsThe repository documents support for these agents. The skills may also work with other agents that can load SKILL.md files, but they may need some setup or small changes.
Want to install it?
Claude Marketplace · Claude Plugin · Manual CopyRepository instructions