OpenAI’s Codex gets reusable cloud environments that follow developers across devices
OpenAI Group PBC today rolled out reusable cloud environments for its Codex coding agent, giving developers a saved project configuration they can work from across devices. Announced at the company’s DevDay conference, the release targets the preparation that comes before an agent writes a line of code. A cloud environment does that work once. Each […] The post OpenAI’s Codex gets reusable cloud…
OpenAI has introduced reusable cloud environments for its Codex coding agent, allowing developers to save project configurations that can be accessed from any device. This feature was unveiled at the company's DevDay conference and is designed to streamline the preparation phase before Codex generates code. Once a cloud environment is set up, subsequent tasks commence from this saved setup, with all required dependencies pre-installed.
Developers initiate the process by linking a GitHub repository, after which Codex analyzes the code to identify project dependencies and generates the necessary installation scripts. Any modifications made during the development process remain within the task, enabling developers to reopen the task later from various devices. These environments are also beneficial for teams, as shared access enables multiple developers to work from the same approved configuration without altering the original setup.
Security is ensured through standard environment variables and network secrets, where the latter only reveal the actual values to the intended programs. Additionally, Codex can access common package registries, with the option to expand internet access as needed. This feature is currently available to ChatGPT Plus, Pro, Business, Enterprise, Edu, and Healthcare subscribers.
However, it is important to note that Codex does not support repositories on GitLab or self-hosted GitHub Enterprise Server, and it is currently limited to working with tasks initiated from a computer or browser.
Written by urgent.news from SiliconANGLE's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.