OpenAI expands Codex with desktop & workflow tools
OpenAI has updated Codex with broader desktop and workflow features for software developers, targeting the more than 3 million people who use the tool each week.
The update expands Codex beyond code generation into desktop control, browser-based work, image creation and longer-running task management. It also adds support for reviewing pull requests, handling GitHub review comments, opening multiple files and terminal tabs, and connecting to remote development boxes over SSH in alpha.
Desktop reach
One of the main additions is background computer use on Mac. Codex can now interact with applications on a user's computer by seeing, clicking and typing with its own cursor, while several agents run in parallel without disrupting work in other apps.
That gives developers a way to use the tool in workflows that are not limited to an application programming interface, including testing apps, iterating on front-end changes and working inside software that does not expose an API.
Codex is also expanding into web-based tasks through an in-app browser. Users can comment directly on pages to guide the agent, a feature currently useful for front-end and game development.
Broader tooling
The release includes more than 90 additional plugins, combining app integrations, skills and model context protocol servers. Named additions include Atlassian Rovo, CircleCI, CodeRabbit, GitLab Issues, Microsoft Suite, Neon by Databricks, Remotion, Render and Superpowers.
Inside the app, users can now open files directly in a sidebar and preview PDFs, spreadsheets, slides and documents. A new summary pane shows an agent's plans, sources and outputs in one place.
Codex can also use gpt-image-1.5 to create and refine images. The feature works with screenshots and code to produce product concepts, design mock-ups and game assets within the same workflow.
Longer tasks
Another part of the update focuses on retaining context over time. Automations can now re-use existing conversation threads, allowing Codex to preserve context from previous work rather than starting afresh with each task.
OpenAI has also added scheduling, letting Codex set future work for itself and resume a task automatically. That could allow work to continue over days or weeks. Teams already use automations for tasks such as progressing open pull requests, following up work items and monitoring conversations across Slack, Gmail and Notion.
A preview of memory is also being introduced. The feature lets Codex retain information from earlier interactions, including user preferences, corrections and details that took time to gather.
OpenAI is also introducing context-aware suggestions. Using project information, connected plugins and memory, Codex can suggest actions to help users resume work, including identifying open comments in Google Docs and pulling related context from Slack, Notion and a codebase.
Rollout
The updated features are rolling out to users of the Codex desktop app who sign in with ChatGPT. Personalisation functions, including memory and context-aware suggestions, will reach Enterprise, Edu, and EU and UK users later. Computer use is initially limited to macOS and will also be extended to EU and UK users later.
The changes reflect a broader shift in how coding assistants are being positioned in software development. Rather than serving only as tools for writing snippets of code, they are increasingly being used to review work, gather context across applications and maintain continuity in projects that unfold over longer periods.
OpenAI said that shift is reflected in how Codex usage has changed since launch. Developers are increasingly using the product not just to write code, but also to understand systems, debug issues, coordinate with colleagues and keep longer-running work moving.
The latest release brings Codex closer to the tools, workflows and decisions involved in building software, with support spanning desktop applications, browser tasks, connected business tools and persistent work threads.