OpenAI Expands Codex with Persistent Cloud Environments and New Developer Tools
Persistent Cloud Development Environments
OpenAI unveiled a set of upgrades for its software engineering assistant Codex, most notably reusable cloud development environments that remain active across sessions and devices. Announced during OpenAI’s DevDay on Tuesday, the feature lets developers launch Codex on a laptop, a phone, or directly in the cloud without starting from scratch each time. The environments retain approved configurations and permissions, giving teams a shared workspace that accelerates task initiation and maintains consistency.
Enhanced Command-Line Interface and Voice Control
The refreshed Codex CLI now supports voice‑driven task creation and management, allowing developers to issue commands hands‑free. A new /agents view centralizes the delegation of work and provides real‑time tracking of multiple concurrent tasks. Additional refinements streamline everyday actions such as editing prompts, resuming interrupted sessions, and navigating worktrees. A cleaner terminal user interface further improves readability during extended use.
Integrated Code Review in ChatGPT
Codex has been woven into a new code review experience inside the ChatGPT desktop application. Users can read concise summaries, explore proposed changes, and query Codex about potential issues before submitting feedback on GitHub pull requests or GitLab merge requests. An automatic review feature also lets Codex perform an initial assessment while the developer steps away from the computer.
Security Enhancements with Codex Security Cloud
OpenAI introduced Codex Security Cloud, a suite of tools aimed at hardening infrastructure. The system can scan entire GitHub repositories on demand, on a schedule, or whenever new commits are pushed. It then investigates findings, eliminates duplicates, and prepares fixes in the cloud, even when the user’s laptop is closed. The offering includes access to the models from OpenAI’s cybersecurity initiative Daybreak Blue without requiring a separate application.
API Updates and New Decision‑Making Capabilities
Among the API enhancements is a new Decisions API that leverages Luna to answer user‑defined questions with predefined responses, enabling real‑time decision support. An updated Agents API now supports computer use and integrates with Amazon’s Bedrock Managed Agents, allowing developers to build OpenAI agents that run entirely on AWS.








