Cursor iOS App Adds Remote Control for Local Agents
Cursor’s new iOS feature lets developers monitor and message agents running on their own computers, extending coding workflows beyond the desktop.
What changed
Cursor has added remote control for local coding agents to its iOS app. Developers can see which agents are running on a paired computer, inspect their current work, and send follow-up instructions from an iPhone or iPad. The agents themselves continue running on the user’s computer rather than being moved into Cursor’s cloud.
The feature is enabled by default for most users, while Enterprise organizations must activate it through administrative settings. Cursor says the computer must remain powered on and connected to the internet; users can also configure the desktop app to stay awake while remote control is active.
Why it matters
This is a small interface change with a larger strategic implication. Coding agents are moving from interactive assistants that wait beside a developer to background workers that can continue across time and place. Mobile access turns the phone into a monitoring and intervention layer rather than a full development environment.
Keeping execution local also gives Cursor a differentiated security and privacy story. Source files, terminals, credentials, and development tools remain on the workstation, but the convenience of remote supervision becomes available away from the desk. That combination may appeal to teams that want agent autonomy without sending every workflow to a hosted environment.
The limitation is operational: a laptop that sleeps, loses connectivity, or is closed cannot host the agent. Cursor’s mobile app therefore extends local agents; it does not make them continuously available in the way cloud agents are. The product’s significance will depend on whether developers trust mobile prompts enough to let long-running agents work unattended.