OpenAI rolls out Chronicle, which builds memories from screen captures to make Codex more aware of context, as a research preview for Pro subscribers on macOS
Last week, OpenAI released an all-new version of Codex for Mac that includes the best example of AI-driven computer use yet.
9to5MacZac Hall
Context & Ripple Effects
OpenAI’s Codex coverage has moved quickly from a macOS command center for managing coding agents to a desktop app with computer control, browsing, automation memory, plugins, and image generation. Chronicle extends that trajectory by making prior on-screen activity available as context rather than treating each agent session as isolated.
The change matters because Codex’s desktop surface is becoming both the place where agents act and the place where they accumulate workflow state. Chronicle is initially constrained to a macOS research preview for Pro subscribers.
First-order effects
Pro subscribers on macOS can test a Codex capability that derives reusable context from screen captures, potentially reducing the need to restate recent work to the agent.
OpenAI adds persistent, screen-level context to a Codex app already expanding beyond code generation into computer control and automation.
Second-order effects
Competing desktop coding-agent products face pressure to improve continuity across browser, app, and coding sessions—not just task-level model performance.
Capturing and reusing screen-derived context raises the importance of clear memory controls and boundaries around what an agent retains, particularly as computer-control features broaden the range of visible work.
Third-order effects
If this approach proves useful, agent platforms may differentiate less through single prompts and more through their ability to maintain governed, workflow-specific context across a user’s work surface.
The desktop AI client could increasingly become a consolidated agent workspace, combining execution, browsing, automation, and accumulated memory; user trust in how that memory is handled will shape adoption.
The trend: This is part of the shift from standalone coding assistants toward persistent, context-engineered agent workspaces that can observe and act across everyday desktop workflows.
Last week, we released a preview of memories in Codex. Today, we're expanding the experiment with Chronicle, which improves memories using recent screen context. Now, Codex can help with what you've been working on without you restating context. [video]
We are releasing a *research preview* of Chronicle in Codex. It allows codex to build up memories based on your day to day work on your computer and then refer to these memories to be a lot more helpful. Available for PRO subscriptions and on Mac to start. This is early and co…
the Codex x @skybysoftware acquisition may have been one of the best @openai deals made in the last year. I've been waiting for “real” computer use since @romainhuet demoed the ChatGPT App with 4o Vision at AIEWF 2024... and only now it's really, actually rolling out in a usable
Chronicle is an experimental feature giving Codex the ability to see and have recent memory over what you see, automatically giving it full context on what you're doing. Feels surprisingly magical to use.
With Chronicle, Codex can better understand what you mean by “this” or “that.” Like an error on screen, a doc you have open, or that “thing” you were working on two weeks ago. Over time, it helps Codex learn how you work: the tools you use, the projects you return to, and the
Chronicle runs background agents to build memories from screen captures. It uses rate limits quickly. Screen captures are stored temporarily on device to generate memories—also stored on device. You can inspect and edit memories. Be aware that other apps may access these files.
Another low key research preview! Codex has memories, and now they can include recent desktop context—ie what you've been working on I've been excited to ship something like this since I joined @OpenAI to work on desktop Super experimental. Pro only for now. Share feedback 🙏