autoresearch-skill
Automatically test and improve any AI skill prompt
All Skills in This Repository (2)
Pure Emerald Level Indicatorsautoresearch
Run autonomous experiments to iteratively optimize a SKILL.md against binary evaluation criteria.
autoresearch
Optimizes Claude Code skills by running eval-scored experiments and mutating prompts autonomously.
Frequently Asked Questions
FAQPage SchemaHow to install autoresearch-skill?โผ
Run `npx skills add olelehmann1337/autoresearch-skill --all -g -y` in your terminal to install it globally.
How to automatically improve an AI skill prompt?โผ
Point autoresearch at your skill, define 3-6 yes/no quality checks, and it runs an autonomous loop that tests, mutates, and keeps only improvements.
What makes a good eval for skill testing?โผ
Every eval must be a binary yes/no question that is specific and checkable, like 'Does the code run without errors?' instead of vague scales or subjective ratings.
Does autoresearch work with Claude Code?โผ
Yes. It is built as a standard SKILL.md file and runs directly inside Claude Code and other compatible AI agents.
Can I watch the optimization progress live?โผ
Yes. The skill generates a self-updating HTML dashboard showing score charts, experiment history, and per-eval breakdowns in your browser.
Related Repositories in Software Engineering
View All in Software Engineeringโopenclaw
Run a personal AI assistant across your devices and chat apps
superpowers
Gives coding agents a disciplined workflow from idea to merged code
react
AI agent skills for building, testing, and porting React core