OpenAI launches dots, an always-on cloud agent for dev work
- Source
- x.com
- Date
dots signals where OpenAI wants background coding to go. The agent runs continuously on its own cloud computer with plugin access. OpenAI describes it triaging bugs, scoping failing builds, and handing work to Codex, which returns complete pull requests for a human to review.
The supporting infrastructure shipped alongside it. Codex cloud environments are reusable repo-and-dependency setups that keep running tasks with the laptop closed, can be steered from web or mobile, and can be shared across Business and Enterprise workspaces. Codex and the ChatGPT Work desktop app also gained in-app PR review with diff inspection, and ChatGPT adds support for the proposed MCP Events specification, so events in connected apps can start automations.
Taken together, the pieces describe a loop where an event triggers an agent, the agent produces a PR, and a person reviews it. The review step is where teams will need to put their attention.
dots. An always-on agent with its own cloud computer and access to your plugins and context. Powered by GPT-6 Astra, your dot proactively keeps your projects moving. Here’s what developers can hand off to their dot: – Triage recurring bug reports and feature requests across your connected apps. – Scope failing builds and feature improvements around your priorities. – Build and test changes with Codex, then bring back complete PRs for your review. Open your dot’s cloud computer to inspect the work, or connect your own so it can work with your local files and code.
Context
OpenAI describes dots as agents that keep working on a developer's projects without waiting to be asked. In the announcement from OpenAI Developers, a dot is an always-on , powered by GPT-6 Astra, with its own cloud computer and access to the developer's plugins and , and OpenAI says it proactively keeps projects moving.
The work OpenAI lists covers a task from intake to a finished change. A dot can triage recurring bug reports and feature requests across connected apps, scope failing builds and feature improvements around the developer's priorities, and then build and test changes with Codex, OpenAI's coding agent, before bringing back complete pull requests, the proposed code changes a developer reviews before merging. That review is where the post places the developer's decision: the agent prepares the change, and the person decides whether to accept it.
The cloud computer is the dot's own workspace. A developer can open it to inspect the work, or connect their own computer so the dot can work with local files and code. The post does not cover pricing, availability or what a dot may do without approval before the pull request, and its attached video had no transcript. Another announcement of the same launch is covered in OpenAI launches dots: always-on agents with their own computer.
- AI agent — An AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.
- context window — The maximum amount of text a model can consider at once — its working memory for the current conversation or task.
Checking sign-in…
Loading comments…