Andrej-Karpathy-s-Autoresearch-As-a-Universal-Skill
Self-improving prompt optimization through autonomous experiment loops
All Skills in This Repository (1)
Pure Emerald Level IndicatorsFrequently Asked Questions
FAQPage SchemaHow to install Autoresearch Universal?▼
Run `npx skills add balukosuri/Andrej-Karpathy-s-Autoresearch-As-a-Universal-Skill --all -g -y` in your terminal to install it globally.
What does Autoresearch Universal do?▼
It applies Karpathy's autoresearch pattern to any repository: it scans your code, picks measurable quality targets, and runs a self-improving loop that keeps better prompts and discards worse ones.
What can I optimize with this skill?▼
Anything measurable, such as test coverage, docstring quality, accessibility, error handling, or documentation SEO. The skill suggests targets based on your repo or accepts your own goal.
Does it work with Cursor and Claude Code?▼
Yes. The SKILL.md file follows the universal skill standard and runs in Cursor, Claude Code, and other compatible coding environments.
Do I need machine learning experience to use it?▼
No. You simply describe what you want to improve in plain English, review the proposed metrics, and let the loop run unattended.
Related Repositories in Software Engineering
View All in Software Engineering→
openclawopenclaw
Run a personal AI assistant across your devices and chat apps
obrasuperpowers
Gives coding agents a disciplined workflow from idea to merged code
facebookreact
AI agent skills for building, testing, and porting React core