Master Your GitHub Copilot Sessions from Any Device with Remote Control
Developers often juggle multiple coding sessions—refactoring in VS Code, debugging via CLI, and scaffolding new features in the background. Until recently, monitoring and steering these workflows required being at your desk. Now, GitHub Copilot remote control lets you take your sessions anywhere, with full visibility and control on github.com, the GitHub Mobile app, VS Code, and JetBrains IDEs. This feature, now generally available for Copilot CLI sessions, creates a seamless multi-surface experience. Below, we answer common questions about how it works and how to get started.
What is remote control for GitHub Copilot and how does it work?
Remote control is a feature that extends GitHub Copilot sessions beyond your local machine. Using the /remote on command, you can transfer an active session—from VS Code or the CLI—to the web or the GitHub Mobile app. This allows you to monitor progress, send follow-up instructions, and approve permission requests from any device. It works with any repository and even directories without repos, so setup isn't a barrier. The technology ensures that what Copilot is doing—the plans it researches, files it reads, changes it makes, and commands it runs—is visible in real time, regardless of your location. This creates a continuous, uninterrupted workflow across CLI, IDE, web, and mobile, making it easier to manage multiple agents simultaneously.

How can I start using remote control for my Copilot sessions?
To begin, you need the latest version of GitHub Copilot CLI or GitHub Copilot in VS Code. If you already have them installed, there's nothing extra to download. Simply start a session as usual—whether in the CLI or VS Code—and type /remote on. Your session will instantly become available on github.com and the GitHub Mobile app. For first-time CLI users, install the GitHub Copilot CLI via the official installation guide. Once set up, you can initiate a session and use the command to send it to your mobile or web interface. The process is straightforward: no configuration changes, no additional accounts. This ease of access lets you step away from your desk without losing any momentum.
Can I monitor and redirect a running Copilot session from my phone?
Absolutely. With remote control, you can open a session on any device, including your phone via the GitHub Mobile app, and watch Copilot's progress in real time. You'll see detailed activity—what plans it's researching, which files it's reading, the changes it's making, and the commands it's executing. If the agent is veering off course, you can send a follow-up instruction in natural language to redirect it. For example, you might tell it to focus on a specific file or adjust its approach. You can also approve or deny permission requests on the go. This capability turns your phone into a powerful command center, letting you steer complex workflows even when you're away from your primary machine.
What is the complete workflow possible with remote control?
Remote control supports an entire developer lifecycle once a session is moved to the web or mobile. A typical workflow might start with /planand scaffold using Copilot CLI. Then you use /remote on to shift monitoring to your phone or web browser. From there, you can send follow-up instructions to steer the session, review the implementation plan and proposed changes, and even create and review a pull request—all from your mobile device. Finally, you merge and move on without missing a beat. This end-to-end capability removes the friction of switching surfaces, whether you're reviewing code on a tablet or approving changes on a phone.

Is my work private when using remote control?
Yes, privacy is built into the design. Your remote sessions are only visible to you. No other users can see or access your sessions, even when they're transferred to the web or mobile. The feature maintains full privacy by default, ensuring that your code, instructions, and Copilot's actions remain confidential. This is crucial for developers working on sensitive projects or in team environments where code privacy is paramount. The security model aligns with GitHub's existing safeguards, so you can use remote control with confidence, knowing your work stays private.
How does remote control support developers working across multiple surfaces?
Remote control is designed to unify your development experience across all major surfaces. You can start a session in VS Code, a CLI, or a JetBrains IDE, and then transfer it to the web (github.com) or the GitHub Mobile app using the /remote on command. This means you can switch devices without interrupting your workflow. For instance, you might begin a complex refactoring task on your desktop, then monitor its progress from a tablet during a meeting, and later approve a pull request from your phone while commuting. The feature works with any repository and directory, so no matter how your project is structured, you maintain continuity. This multi-surface support reduces the pain of context-switching and lets you stay productive wherever you are.
What do I need to get started with remote control?
Getting started is simple. If you already use the latest version of GitHub Copilot CLI or GitHub Copilot in VS Code, you don't need to install anything new. Just start a session and enter /remote on. For those new to Copilot CLI, install it from GitHub's official instructions. There are no additional permissions or configuration steps. Once set up, your sessions appear on github.com and the GitHub Mobile app automatically. The feature is available now for all users with Copilot access. For detailed instructions, refer to the official GitHub documentation. Remote control turns your development into a seamlessly portable experience, empowering you to manage agentic workflows from any location.
Related Articles
- The Rise of Enterprise Vibe Coding: Unprecedented Productivity Meets AI Governance Challenges
- Inside Python’s Security Response Team: Governance, Growth, and How to Join
- Go 1.26 Introduces Rewritten `go fix` for Automatic Code Modernization
- Python 3.15.0 Alpha 5 Unveiled: New Features and Performance Boost
- Frontend Engineers Face New Crisis: Microservices Complexity Threatens User Experience
- Go 1.26 Launches Revamped 'go fix' to Automate Code Modernization
- OpenClaw AI Agent Sparks Security Crisis: Mass Deletion Incident Exposes Risks of Autonomous Assistants
- Pyroscope 2.0 Q&A: Everything You Need to Know About Next-Gen Continuous Profiling