ARTIFICE DAILY AI + CYBER SECURITY BRIEFING

Artificial Intelligence

OpenAI adds cloud environments and review tools to Codex

OpenAI announced reusable cloud environments, a refreshed Codex CLI, new review features and security tools for Codex, alongside API changes that extended its agent system to AWS.

TechCrunch

OpenAI said Codex had gained reusable cloud development environments that could be reached from any device. It said those environments were intended to start work faster and give teams a shared workspace with approved settings and permissions.

The company also updated the Codex CLI so developers could start and direct tasks using their voice. OpenAI added a /agents view, plus changes for editing prompts, resuming sessions, using worktrees and a cleaner terminal interface.

Codex also appeared in a new code review experience inside the ChatGPT desktop app. OpenAI said users could read summaries, inspect changes or ask Codex about issues before commenting on GitHub pull requests or GitLab merge requests, and that automatic reviews could run while users were away.

OpenAI introduced Codex Security Cloud for scanning repositories on demand, on a schedule or when new commits arrived. It said the system could investigate findings, remove duplicates and prepare fixes in the cloud, and that it included models from Daybreak Blue without a separate application.

The company also announced a Decisions API that used Luna for predefined answers. It updated the Agents API to support computer use and Amazon’s Bedrock Managed Agents, allowing agents to run entirely on AWS.

Named in this story

Companies

OpenAI
announced the Codex updates and API changes
AWS
platform where OpenAI agents could run

Organisations

Daybreak Blue
cybersecurity initiative whose models were included

Products and systems

Codex
software engineering agent gaining the new features
Codex CLI
command-line tool updated with voice control and other changes
ChatGPT desktop app
hosted the new code review experience
Codex Security Cloud
toolset for scanning repositories and preparing fixes
Decisions API
new API for real-time decisions
Luna
model used by the Decisions API
Agents API
updated API that added computer use support

How the source tells it

Mostly neutral product reporting, with a mild boosterish tone created by OpenAI’s own benefit-led descriptions of convenience and security.

  • vendor boosterism OpenAI’s own benefit claims are repeated without outside validation