Cursor launches a native iOS app for controlling coding agents from your phone

Cursor released a native iOS app in public beta, letting developers launch and steer AI coding agents from their phone instead of only from a desktop editor.

What's new

Cursor for iOS is available now in public beta on all paid plans. As the company describes it, "With Cursor for iOS, you can launch always-on agents in the cloud, or control agents running on your computer from your phone." The app covers three core workflows:

• Launching cloud-based agents that run in isolated virtual machines with full development environments — Cursor says "Cloud agents run in isolated virtual machines with full development environments to test, verify, and demo work" — so an agent can pull a repo, make changes, and produce a working build without touching the developer's own machine.
• Remotely monitoring and steering agents already running on a developer's own computer, with push notifications when work is ready for review.
• Reviewing diffs, resolving comments, and merging pull requests directly from the app once an agent finishes a task.

Cursor frames the use case around interruptions that don't wait for a laptop: "When you get paged at lunch, you can kick off an agent to investigate and propose a fix," and similarly for a customer bug reported while a developer is away from their desk. Voice input and slash commands let users describe a task or steer an agent without typing much on a phone screen, and any supported frontier model can be selected for a session. As a launch promotion, Cursor is offering 75% off Composer 2.5 runs inside the mobile app through July 5, 2026.

Context

The release follows a broader industry move to make coding agents accessible outside the IDE — GitHub Copilot, Replit, and others have pushed mobile or web-based agent monitoring over the past year as "agent that runs unattended for hours" becomes a more common pattern than "autocomplete inside an editor." Cursor itself has leaned heavily into cloud/background agents in recent releases; the iOS app is the natural extension of that: if an agent can run unattended in the cloud, someone eventually needs a way to check on it without being at a desk.

Why it matters

As coding agents take on longer-running, more autonomous tasks — refactors, bug investigations, full feature branches — the bottleneck increasingly shifts from "can the agent do the work" to "can a human review and approve it fast enough to keep the loop moving." A mobile app that supports full review-and-merge, not just monitoring, treats the phone as a legitimate control surface for software delivery rather than a notification-only companion. For a tool already popular with individual developers and engineering teams, it also lowers the cost of running agents overnight or over a weekend, since someone can approve or redirect work from anywhere rather than waiting to get back to a desk.