Skip to content

Upgraded Codex developer agent from OpenAI receives major modifications

AI giant OpenAI expands accessibility to its coding assistant, Codex, incorporating platforms such as GitHub and Integrated Development Environments (IDEs), broadening its reach in the programming world.

AI's Codex development tool has received a significant revision
AI's Codex development tool has received a significant revision

Upgraded Codex developer agent from OpenAI receives major modifications

OpenAI has announced a significant update to its AI coding agent, Codex, making it more accessible and powerful than ever. The update includes a refreshed user interface, new commands, bug fixes, and the addition of code reviews in GitHub, all aimed at streamlining the coding experience.

One of the key features of the update is the ability for developers to delegate tasks to the cloud to execute asynchronously without losing state. This feature is available not only for Codex users but also for developers using popular code editors like VS Code, Cursor, and Windsurf. The Codex CLI tool, which can be installed using package managers like npm or Homebrew, enables local execution of Codex on users' computers.

The update also sees Codex becoming more integrated into developers' workflows. Codex is now accessible from a terminal, IDE, web, GitHub, and even from the ChatGPT iOS app. Developers can sign in with ChatGPT via both IDE and command-line interface (CLI), eliminating the need for API key setup and utilizing existing ChatGPT subscriptions.

OpenAI has also introduced an Integrated Development Environment (IDE) extension for Codex, expanding the availability of Codex for IDEs primarily through plugins and installations. This move is part of OpenAI's ongoing efforts to make Codex more accessible and versatile.

The updates come approximately three months after the initial release of the cloud-based Codex AI agent. This update is a response to the growing competition in the AI coding field, with AWS, Google, Microsoft, and GitHub Copilot all extending their AI coding capabilities this year.

However, the path hasn't been without challenges. OpenAI's recent release of GPT-5 sparked a backlash due to issues with accuracy and tone of voice, as well as the initial removal of access to GPT-4. Despite these setbacks, OpenAI continues to push the boundaries of AI technology.

A recent survey suggests that AI is currently better at coding than most humans, though issues around code quality persist. To address these concerns, OpenAI has directed Codex users to automatically review new pull requests (PRs) in a repo or mention @codex in PRs to get reviews and suggested fixes.

Before the update, Codex was only accessible within ChatGPT via a sidebar in the chatbot. The update marks a significant step forward in OpenAI's mission to make AI a more integral part of the coding process, making it easier and more efficient for developers to write and review code.

OpenAI has provided a dedicated website for future Codex updates, ensuring that developers stay informed about the latest features and improvements. With these updates, OpenAI continues to demonstrate its commitment to advancing AI technology and making it more accessible to developers worldwide.

Read also:

Latest