OpenAI Adds Persistent Cloud Environments, Security Scanning, and Voice Controls to Codex
Codex now offers reusable cloud development environments that persist across sessions and work from any device, including phones, with shared settings and permissions. Its CLI also adds voice controls and an /agents view for managing parallel tasks.
OpenAI announced a broad set of updates to Codex, its software engineering agent, at the company’s Dev Day on September 29, 2026. The changes center on making Codex more accessible across devices and adding dedicated security tooling for developers.
The headline addition is reusable cloud development environments that persist between sessions and can be accessed from any device — including phones — rather than treating each cloud task as a temporary, isolated sandbox. OpenAI says the environments are designed to help tasks start faster and give teams a shared workspace with approved settings and permissions.
The Codex CLI is also receiving an update that adds voice controls, allowing developers to start and direct tasks by speaking. A new /agents view makes it easier to delegate and track multiple tasks simultaneously. Other CLI improvements include easier prompt editing, session resuming, worktree support, and a redesigned terminal interface intended to make longer sessions more readable.
Codex is also being integrated into a code review experience within the ChatGPT desktop app. Developers can read summaries, explore code changes, and ask Codex about potential issues before submitting feedback on GitHub pull requests or GitLab merge requests. An automatic review mode lets Codex conduct an initial review while the developer is away.
On the security front, OpenAI introduced Codex Security Cloud, a set of tools that can scan entire GitHub repositories on demand, on a schedule, or when new commits arrive. Codex then investigates findings, removes duplicates, and prepares fixes — even when the developer’s laptop is offline. The toolset includes access to models from OpenAI’s cybersecurity initiative Daybreak Blue without requiring a separate application.
OpenAI also announced several API updates, including a new Decisions API for real-time decision-making using a model called Luna, and an updated Agents API that now supports computer use and integration with Amazon’s Bedrock Managed Agents, enabling OpenAI agents to run entirely on AWS.