leo-lilinxiao/codex-autoresearch
An autonomous iterative system implemented as an OpenAI Codex skill that cycles through modify, verify, and retain operations on code.
Not currently ranked — collecting fresh signals.
star history
The repository provides a Codex skill that enables autonomous goal-driven experimentation. Users specify what they want to improve, and the system uses Codex to modify code, run verification commands, decide whether to keep or discard changes, and repeat the loop indefinitely. It supports mechanical verification of any command-defined metric including test coverage, type errors, latency, lint warnings, and security findings.
Frequently asked
- What is leo-lilinxiao/codex-autoresearch?
- An autonomous iterative system implemented as an OpenAI Codex skill that cycles through modify, verify, and retain operations on code.
- Is codex-autoresearch open source?
- Yes — leo-lilinxiao/codex-autoresearch is open source, released under the MIT license.
- What language is codex-autoresearch written in?
- leo-lilinxiao/codex-autoresearch is primarily written in Python.
- How popular is codex-autoresearch?
- leo-lilinxiao/codex-autoresearch has 2k stars on GitHub.
- Where can I find codex-autoresearch?
- leo-lilinxiao/codex-autoresearch is on GitHub at https://github.com/leo-lilinxiao/codex-autoresearch.