On 2026-10-23 (15:59 UTC) the official Cursor account @cursor_ai announced "Cursor can now control your computer": agents can drive desktop apps such as the iOS Simulator to test the features they build. A follow-up post says Computer Use lets agents click and type across any app on your computer, not just the browser. Cursor turns it on by itself when a task needs it, and you can also request it by @-mentioning Computer Use in your prompt; updating Cursor is enough to try it. A community changelog in September mentioned Computer Use on Mac in version 3.20.21, and this is the official launch. Supported platforms, the permission model and benchmarks were not published.
Key Takeaways
- ✓Cursor officially says its agents can now control your computer and drive desktop apps like the iOS Simulator to test the features they build
- ✓Computer Use lets agents click and type across any app, not just the browser
- ✓Cursor uses it automatically when a task needs it, and you can request it by @-mentioning Computer Use in a prompt
- ✓Update Cursor to try it; supported platforms, the permission model and benchmarks were not published
Key Decision Metrics at a Glance
Turn your technical choice into a development budget
Compare 40 dev plans & simulate token costs vs $20/mo subscriptions
Project Links & Resources
Direct AccessIn-Depth Technical Analysis
Background and pain point
After a coding agent builds a feature, verification often stalls at the last step: browser tools can only test web pages, so iOS and desktop apps that need a real GUI still have to be clicked through by hand. On October 23 (15:59 UTC) Cursor officially announced "Cursor can now control your computer", extending agent verification from the browser to the whole machine.
How it works
According to Cursor, Computer Use lets agents click and type across any app on your computer, not just the browser. In the demo video an agent drives the iOS Simulator to test a feature it just built. There are two ways it starts: Cursor uses it on its own when a task needs it, and developers can ask for it by @-mentioning Computer Use in the prompt. Cursor has not described the underlying implementation, supported operating systems or permission controls.
Benchmarks
Not published. Cursor gave no task success rates, speed figures or comparisons with other computer-control tools.
What developers can do now
Update Cursor to the latest version to try it. Mobile and desktop teams can have the agent test its own work in the iOS Simulator and then review the result themselves. Because the agent can operate any app on the machine, start in a clean test environment, review the system permissions you grant, and keep windows with sensitive data out of its reach.
Benchmark side-by-side against alternatives, or calculate monthly token cost vs subscription break-even.
Discussion & Comments
0Sign in to join the discussion
Connect with AI developers to exchange benchmark insights.