Cursor: Remote Control for Local Agents, Plus Rollouts and Security Review Bots
Cursor's iOS app can now show and reply to agents running on your own computer (6 October 2026), without cloud agents. Earlier, on 23 September, Cursor launched Rollouts, which monitors a pull request from merge to production and flags regressions, and Security Review, which posts exploitable-bug findings on each PR, for Teams and Enterprise.
Cursor shipped three features between late September and early October 2026 aimed at what happens around the code: watching agents from your phone, and checking changes after they're written.
Remote control for local agents (6 October)
The Cursor iOS app can now display and reply to agents running on your computer — no cloud agents required.
- It's on by default, except for Enterprise organisations, where admins can enable it.
- Your computer must stay on and online; the agent still runs there.
This is the same idea as Claude Code's Remote Control, which lets you continue a local session from the Claude app on your phone. The limitation is shared too: close the laptop lid, and the agent stops. (Claude Code From Your Phone)
Rollouts (23 September, Teams and Enterprise)
Rollouts attaches a monitor to each pull request and tracks its health in each environment during deployment, flagging regressions. Cursor included temporary usage credits (enough for roughly 50 changes on Teams or 500 on Enterprise over 10 days) so teams could try it.
It's an AI take on a classic practice: deploy gradually, watch error rates and key metrics, and roll back quickly. (Blue-Green vs Canary Deployments, Feature Flags)
Security Review (23 September, Teams and Enterprise)
Security Review posts one review per pull request reporting exploitable bugs, with severity and proposed fixes. Similar features exist across the industry — Claude Code has /security-review and Code Review, and OpenAI announced Codex Security Cloud at DevDay. (Claude Code Security Review, OpenAI DevDay 2026: Codex)
Automated reviewers catch real issues, but they're a layer, not a replacement for tests, dependency scanning and a human reading the diff. (OWASP Top 10 Explained)
Also in Cursor this month
- An agent harness rework that Cursor says cut token costs by about 7%.
- Projects (multi-agent, beta) and self-hosted machines. (Cursor Projects and Self-Hosted Machines)
Source: Cursor changelog
EasySpawn runs Claude Code on a server that doesn't sleep, so you can check in from your phone and the agent keeps working whether your laptop is open or not. See how it works or join the waitlist.
Related: What Is Cursor? · Claude Code vs Cursor · Know When Your App Is Down · Zero-Downtime Deploys
Keep reading
Cursor Adds Projects (Multi-Agent, Beta) and Self-Hosted Machines
In September 2026 Cursor launched self-hosted machines (2 Sept), which keep agent tool execution on your own laptops, VMs or sandboxes, and Projects (10 Sept, beta), where a coordinator agent plans long-running work and delegates it to parallel agents that can react to Slack, schedules and pull requests.
What Is Windsurf? The AI Code Editor, Now Called Devin Desktop
Windsurf was an AI-first code editor built on VS Code, known for its Cascade agent and Tab completions. Cognition bought it in 2025 and renamed it Devin Desktop in June 2026. What it is, what changed, how it compares with Cursor and Claude Code, and what existing users should know.