OpenAI’s CLI agent. We’re building a parser for sandboxed execution logs, file changes, and command outputs.
When it ships
Sandboxed execution sessions, start to finish — what the agent ran and why.
Every file the agent created, edited, or deleted, against what it claimed.
Shell commands and their real outputs, including the ones that failed.
Want Codex support? Be first to know when it ships — every engine is ready the day it lands.
Get notified →Join the waitlist · We notify you when it ships