
OpenAI's Codex can now watch you work once and repeat the task forever
Quick Answer
OpenAI has introduced a 'Record & Replay' feature in its Codex app for macOS, allowing users to demonstrate workflows that Codex can then automate indefinitely.
Quick Take
Currently, this feature is not available in the EU, UK, or Switzerland.
Key Points
- Codex converts demonstrated workflows into reusable 'skills' for automation.
- The feature enhances productivity by allowing Codex to repeat tasks autonomously.
- Currently unavailable in the EU, UK, or Switzerland, limiting access for users in those regions.
- This update signifies a step towards more advanced AI-assisted workflows.
📖 Reader Mode
~1 min readOpenAI released a new feature called Record & Replay for the Codex app on macOS. Users can walk the AI agent through a workflow onc, like uploading a YouTube video with metadata, a thumbnail, and subtitles. Codex turns that recording into a reusable "skill" and can then repeat the process on its own. The feature isn't available yet in the EU, the UK, or Switzerland, and it requires Computer Use to be turned on. Computer Use has been available in the EU since June 16.
Also new in version 26.616: bulk actions for the Automations history, plus the ability to hand off threads between a local and remote host to continue tasks on a connected machine. Codex is OpenAI's AI agent for coding and white-collar-work automation. The app is free to download, but you need a paid ChatGPT account to get real use out of it.
— Originally published at the-decoder.com
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from The Decoder
See more →
An AI model programmed nonstop for 19 days on a single MirrorCode task that cost $2,600 to run
Epoch AI's MirrorCode benchmark reveals Claude Opus 4.7 as the leader with a 56% solve rate, reconstructing a 16,000-line toolkit in 14 hours. Despite this, all models tested struggle with the most complex tasks, highlighting limitations in current AI capabilities. The single task consumed $2,600 over 19 days, raising questions about cost-effectiveness in AI development.

