What Unity Developers Learned About Building for Meta Quest at Unite Seoul 2026
Read time: ~8 minutes
Unite Seoul brought together hundreds of Unity developers on July 21 for one of the year's most anticipated gatherings of the global Unity community. We hit the stage with a keynote segment, hosted a dedicated technical deep dive session, and shared hands-on booth demos, all centered on one message: building for Meta Quest inside Unity is faster and more approachable than it's ever been.
Here's what you need to know if you weren't in the room, and what to explore first if you were.
At Unite, attendees lined up for hands-on booth demos and a live look at agentic tools that turn a Unity project into a playable VR prototype in days, not weeks.
On the Keynote Stage: AI Meets VR Development in Unity
Jamie Keane (Director, Product Management, Meta) joined the Unite keynote to lay out how Meta and Unity are collaborating to make Quest development more streamlined from the start.
In short, Unity developers who haven't experimented with VR before can now dive in with confidence and a lower learning curve. AI acts as a knowledgeable companion inside your existing workflow, handling the VR-specific complexity (input systems, performance constraints, platform configuration, etc.) so you can focus on the creative aspects of the experience you want to build.
Keane briefly walked through four live workflow examples on stage:
Project setup and configuration for Quest, with AI-guided scaffolding that handles platform-specific details automatically.
Adding VR capabilities like hand input without requiring tedious setup or hunting through documentation.
Performance optimization, from trace capture through issue identification to recommended fixes, all within the Unity Editor.
Scene understanding and validation, where AI tests interactions autonomously, confirming changes work before you ship.
The takeaway was that these workflows meet you where you are to make building VR for Quest easier and more efficient.
Jamie Keane (Director, Product Management, Meta) demos AI-assisted VR development workflows in Unity on the Unite Seoul keynote stage.
5 Tools That Define the New Quest Development Workflow
The dedicated speaker session dove deep into the tooling stack that makes all of this possible. Here's what matters for your next project:
1. Meta VR CLI: Your Command-Line Entry Point
Meta VR CLI is a combined CLI and Model Context Protocol (MCP) server that is purpose-built for Quest device development. It helps handle environment setup, device and emulator control, performance tooling, and AI-optimized documentation search. It also enables you to connect preferred AI coding assistants (Claude Code, Codex, Copilot, or others) directly to Quest platform tools.
Here's what you can do with it:
Connect to a Quest device and manage apps, builds, and files
Run performance analysis with Perfetto trace tools
Search Meta's developer documentation from the terminal
Compose it with the Unity CLI and Unity MCP server for a full agentic loop
Extra tip:
Meta VR CLI runs in Meta Quest Developer Hub too. If you prefer a GUI starting point, the tooling layers on top without replacing your existing setup.
To get started, learn how to enable AI tools in Unity with Meta VR CLI.
2. Meta XR Operator: Closed-Loop Testing Without Lifting Your Headset
Coming in Meta XR SDK v205, the new Meta XR Operator gives your AI agent full scene understanding and controller input inside Meta XR Simulator. That means your agent can:
Read room geometry (walls, floors, furniture) and reason about 3D space
Inspect the Unity scene hierarchy and components
Take screenshots of the running app as rendered in Meta XR Simulator
Use virtual controllers (press buttons, move thumbsticks, pull triggers, point and aim)
Navigate UI menus and multi-step flows autonomously
Together, these capabilities mean your agent does far more than write code. It sees the running experience, interacts with it, and validates its own work, giving you closed-loop agentic workflows for Quest.
Our agentic tools repository on GitHub contains 32 agent skills that build on Meta VR CLI/MCP and Unity MCP. It contains plugins for Claude Code, Cursor, Codex, VS Code, and Copilot, plus Unity-specific skills for VR project creation, code review, platform integration, API upgrade paths, and Android-to-Quest app porting.
It's also Apache-2.0 licensed, so there's no lock-in and nothing proprietary gating your workflow. Fork it, extend it with your own skills, or plug the existing 32 directly into whatever agentic setup you're already running. If you're using Claude Code, Cursor, Codex, or VS Code with Copilot, the repository includes ready-made plugins that connect in minutes.
4. IDE Extensions: Meta Horizon for VS Code and Android Studio
Whether you're in VS Code, Android Studio, or prompting through Gemini or Copilot, you get device status, documentation search, MCP registration, and build shortcuts without switching windows.
The Meta Horizon extension on the VS Code Marketplace and JetBrains Marketplace bring platform tools, documentation search, and more into your editor.
5. AI-Ready Documentation
Meta's developer documentation is now formatted in AI-friendly markdown (.md) with an AI sitemap per build path (llms.txt). Token-efficient documentation search runs through Meta VR CLI and the MCP server, so your agent gets the right context without blowing through its context window.
Inside Unity: The Tooling Layer That Ties It All Together
Beyond the agentic workflow stack, the session covered major updates to the core Unity development tools shipping in Meta XR SDK v205.
Building Blocks (Now with AI)
Building Blocks let you discover and drop Meta XR features directly into your project in Unity. You can explore pre-built components for immersive experiences, mixed reality, multiplayer, tabletop games and more. Here's what's new with the latest updates:
AI Building Blocks for integrating on-device AI
Passthrough Camera Access Building Blocks for working with the passthrough cameras
With Building Blocks, you can drag in or click on a given feature, have it implemented and optimized for Quest, and start iterating immediately without manual setup.
Get started by exploring Building Blocks features in our documentation.
Meta XR Simulator (Redesigned)
Meta XR Simulator has been rebuilt from the ground up with a new interface optimized for simplicity and reliability. Here are some of its core capabilities:
Test mixed reality with synthetic rooms (office, living room, corridor, and more)
Simulate all major Quest XR features: hands, controllers, Scene API, Passthrough
Capture and replay sessions
Multiplayer testing
Connect a real headset and use its input without putting it on
For developers iterating frequently, this means fewer headset on/off cycles and faster feedback loops on your PC or Mac. Download the All-in-One SDK to get started and visit our documentation for full details on how XR Simulator can help you test your project.
Meta XR Simulator's synthetic environments let you test how your app behaves in different physical spaces without leaving your PC or Mac.
Runtime Optimizer with AI Debugging (Improved in v205)
The Meta Quest Runtime Optimizer now pairs with AI to interpret CPU and GPU timings for you, identify problem GameObjects, Materials, and Assets, and test the potential benefit of "What If" optimizations before you commit to changes. It builds on Unity's profiler workflows, so the integration feels native and enables you to move faster.
Immersive Debugger with AI (New in v205)
Did you know that you can monitor, visualize, troubleshoot, and iterate on your game from inside your headset? The Immersive Debugger enables all of this, now alongside these new AI capabilities:
Troubleshoot issues live while you're in-headset
Modify properties at runtime through natural language
Call other tools (performance analysis, scene inspection) without leaving your experience
Immersive Debugger now enables an AI assistant to inspect, modify, and verify scene properties through conversation while you stay inside your experience.
The result is fewer headset on/off cycles and more time spent iterating where it counts: inside the experience your players will see.
The Business Case: VR Is a Proven Market
Our deep dive session opened with numbers that matter to anyone evaluating whether to build for Quest:
Meta Quest has millions of lifetime users in over 200 countries.
In 2025, more than 300 apps on the Store on Meta Quest generated more than $1 million in revenue each.
More than ten apps generated more than $50 million.
The tools we shared at Unite Seoul are built around the idea that the wave of successful Quest apps won't require years of VR-specific know-how to build, because AI now shoulders the platform setup, device configuration, and performance tuning that used to take months to learn and days to execute. Now, the opportunity is open to any developer willing to try.
From Prompt to Playable: What Rapid Prototyping Looks Like
One of the most compelling demos in the session showed two VR prototypes built almost entirely through AI-assisted agentic workflows:
Arcade Hoops went from idea to playable beta in two days, then evolved into a tabletop mixed reality game less than two days later.
Arcade Hoops demo shown at Unite Seoul.
Neon Reach VR (a hand-tracked arcade game) hit playable prototype in one day, with a polished version ready in three days total.
Neon Reach VR demo shown at Unite Seoul.
Both were built internally at Meta to stress-test the agentic toolchain and demonstrate what's possible when Meta VR CLI, the Unity MCP, and Meta XR Operator work together.
The workflow: describe what you want ("add a new ring type that moves slowly from left to right"), and the agent writes code, builds the scene, and validates the result inside XR Simulator. Then, you can continue to iterate with prompts, apply manual updates and ship it when it feels right.
Worth noting: these workflows handle platform setup, configuration, and known VR patterns well. Novel mechanics and fine-tuned games still benefit from hands-on iteration, but you begin from a working baseline instead of starting from scratch.
On the Booth Floor: Hands-On with Agentic Tools and Hand Tracking
Our booth at Unite Seoul featured two demo stations:
Station 1: Agentic Tools in Unity Attendees could see Meta VR CLI, the MCP server, and Meta XR Operator working together in Unity. We were thrilled to show real use case examples in which an AI agent set up a Quest project, added features via natural language prompts, and validated its own work through scene understanding and controller interaction.
Station 2: Hand Tracking Hands-on demos of hand tracking interactions on Meta Quest hardware, showing how developers can add physical, controller-free interactions to their games and apps with the help of Building Blocks.
Attendees got hands on with the latest hand tracking updates at the Meta booth.
Get Started Today
Install Meta VR CLI and run metavr init to set up agent skills and configure MCP servers.
Grab the Meta XR SDK from the Unity Asset Store (v205 shipping soon).
Clone the agentic tools repo from GitHub (meta-quest/agentic-tools) and plug relevant skills into your preferred AI coding assistant.
Read our Unity guide covering the agentic tools and workflows helping you build faster.
🎯 Action Item:
If you're a Unity developer who's never shipped on Quest before, start with Meta VR CLI and the Unity MCP. Describe what you want to build. Let the agent handle the VR-specific setup. You'll have a running prototype faster than you expect.
Unite Seoul was an incredible opportunity to connect with developers and show firsthand how much easier it is to build for Quest today. Whether this is your first Quest project or your fifteenth, the path from idea to playable VR has never been shorter. Visit the Meta Quest developer documentation to explore the full toolset, or check out our prompting best practices to get faster, more accurate results.