
Cursor released its iOS app in public beta on June 29. You can launch coding agents from your phone or remotely steer agents already running on your desktop.
Cursor released its iOS app as a public beta on June 29 (local time). The core idea: you can spin up a coding agent directly from your phone, or remotely pick up and steer an agent that's already running on your desktop. You can grab it from the App Store or through the TestFlight public beta.
Per Cursor's announcement, the iOS app is open as a public beta on all paid plans. There's no separate signup — just download it from the App Store and log in with your existing account.
Running the Composer 2.5 model also comes with a 75% discount through July 5, according to Cursor's announcement. It looks aimed at lowering the cost barrier so you'll actually test running agents on mobile.
Just like the desktop app, you can choose a repository, pick a frontier model, and kick off an agent. You can hand it instructions by voice or with slash commands.
Cloud agents run inside an isolated virtual machine. That means your session keeps going even after you close the laptop, and according to the official changelog you can also move work you were running locally over to the cloud and continue it there.
The Remote Control feature lets you keep directing an agent that's running on your computer from your phone. To stay connected while you're away from your desk, you can turn on a "keep computer awake" setting.
On Team and Enterprise plans, an admin has to enable this feature first from the Cursor dashboard before you can use it. If you're on a company account, that's something to check.
You can check agent status via Live Activities on the lock screen, and you get push notifications when work finishes, when input is needed, or when something's ready for review. It's built so you don't miss progress even when you're not at your desk.
You can also look over demos, screenshots, logs, and diffs from your phone, leave follow-up instructions, or merge a PR right from the app. The practical use case that lands here is queuing up a long task and just checking the results during a commute or while you're out.
That said, it's hard to judge voice input accuracy or how the mobile experience actually feels from the announcement alone. That's something to gauge by grabbing the beta and using it yourself.
Source: Cursor — Cursor Mobile App for iOS, Build from anywhere with Cursor for iOS
This post summarizes the official announcement and is not sponsored by Cursor in any form.
Original with full infographics and visual structure: https://jessinvestment.com/cursor-ships-an-ios-app-run-coding-agents-from-your-phone/
cursorCursor Automations in 2026: wire event-triggered coding agents to Slack, CI, and...
aiThe specs exist. The AI just can't see them. I've always been the type who builds hobby...
amazonbedrockConnect Claude Code, Cursor and Codex to Amazon Bedrock's new console (2026) Summary. On 5...
aiThere is a weird uncanny valley with LLM-generated UI right now. The code functions perfectly, but if...
aiI went down a rabbit hole this morning reading the late-2025 Juejin AI roundups side by side, and the...
mcpInstall guide and config at curatedmcp.com Zendesk MCP: Let Claude Handle Your Support...
Workflows from the Neura Market marketplace related to this Cursor resource