codex-autoresearch
Autonomous experiment loop that fixes code until targets are met
All Skills in This Repository (2)
Pure Emerald Level Indicatorscodex-autoresearch
Automate iterative Codex experimentation with planning, execution, verification, and logging.
codex-autoresearch
Runs autonomous Git-backed experiment loops that verify numeric metrics and keep or revert each trial.
Frequently Asked Questions
FAQPage SchemaHow to install codex-autoresearch?โผ
Run `npx skills add leo-lilinxiao/codex-autoresearch --all -g -y` in your terminal to install the skill globally.
What does codex-autoresearch do?โผ
It lets Codex run a controlled experiment loop in your Git repo: change one thing, measure a numeric metric, keep improvements, revert failures, and repeat until your target is reached.
What problems can codex-autoresearch fix?โผ
It works for any measurable outcome such as test failures, type errors, code coverage, warnings, latency, binary size, or reproducible security findings.
Is codex-autoresearch safe for my repository?โผ
Yes. Every trial is a Git commit, failed experiments are automatically reverted, and out-of-scope edits or malformed results stop the run with a clear error.
Can codex-autoresearch run overnight in the background?โผ
Yes. It supports a detached background mode where a controller keeps iterating, and you can ask for status, stop, or resume at any time.
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