SimWorld Studio

Vibe code the physical world. Chat with an AI coding agent to build, simulate, and control 3D environments in Unreal Engine 5 — with embodied agent support, real-time pixel streaming, and a full data visualization stack.

SimWorld_Studio_Demo.mp4

Overview

SimWorld Studio is an AI-native 3D scene authoring and embodied agent testbed built on Unreal Engine 5.3. It connects Claude (via Claude Code CLI) to UE through a multi-protocol bridge, letting you:

Build scenes by chat — describe a city block and watch it materialize in real time
Control embodied agents — spawn, command, and observe AI agents navigating the scene
Run multi-agent experiments — testbed with auto-PIE, trajectory recording, collision tracking
Stream the viewport — Pixel Streaming delivers the live UE viewport to any browser

Architecture

┌─────────────────────────────────────────────────────────┐
│                    Browser (React UI)                    │
│  Coding Agent │ Viewport │ Embodied Agent │ Statistics   │
└──────┬────────────┬──────────────┬────────────┬─────────┘
       │ SSE/HTTP   │ WebSocket    │ SSE/HTTP   │
       ▼            ▼              ▼            ▼
┌─────────────────────────────────────────────────────────┐
│              Web Server  (Node.js / Express)             │
│  index.js  │  AgentController  │  MetricsHub  │  SSE    │
└──────┬──────────────┬──────────────────────────────────-┘
       │              │
       ▼              ▼
┌────────────┐  ┌─────────────────────────────────────────┐
│ Claude CLI │  │            Unreal Engine 5.3             │
│ (MCP tools)│  │  UnrealMCP (TCP:55557)  │  UnrealCV      │
│            │──│  spawn / delete / move  │  vget / vset   │
└────────────┘  │  Python scripting       │  camera / obj  │
                └─────────────────────────────────────────-┘
                              │
                ┌─────────────┴──────────────┐
                │  Cirrus Signalling Server   │
                │  Pixel Streaming  :8685/86  │
                └────────────────────────────┘

Component Map

Component	Path	Description
Launch Scripts	`SimWorld-Studio.ps1 / .bat`	One-click launcher: Cirrus → UE → Web Server
Web Server	`web/server/index.js`	Express API, SSE push, asset catalog, metrics
MCP Server	`web/server/mcp-server.js`	Claude ↔ UE tool bridge via TCP
Agent Controller	`web/server/agent-controller.js`	Per-agent sessions, trajectory, collision tracking
Metrics Hub	`web/server/metrics-hub.js`	Time-series data for all agents (collision, speed, turns)
Context Manager	`web/server/context-manager.js`	Scene state (actors, environment, round)
UnrealCV Bridge	`web/server/unreal-bridge.js`	Persistent TCP socket to UCV with FIFO queue + retry
Frontend	`web/src/App.jsx`	React app: chat, viewport, agents, stats, asset browser
UE Plugin	`UE_Project/Plugins/unrealcv/`	Extended UnrealCV with hit tracking, actor camera commands

Features

🏗️ Coding Agent (Scene Generation)

Natural language → 3D scene in real time
125 buildings (BP_Building_01–127), 6 trees, vehicles, street furniture, static meshes
17 marketplace packs discoverable via list_assets() (allow-AI licensed)
Auto session-suffix on actor names prevents cross-map name collision crashes
verify_scene tool: Claude evaluates screenshot and returns PASS/NEEDS_IMPROVEMENT/FAIL

👁️ Live Viewport

Pixel Streaming via Cirrus signalling server — full UE viewport in browser
Click to activate mouse/keyboard input
1920×1080 default, adaptive resolution via MatchViewportRes

🤖 Embodied Agent Panel

Spawn and control Base_Pedestrian / Base_User_Agent / Base_Demo agents
Per-agent session: position, rotation, heading (compass), speed, status
Real-time state: background poller every 3s via UnrealCV
Trajectory view: top-down SVG map with heading arrows, collision markers (red dots)
Camera tab: 3-strategy agent POV (actor camera → camera/0 at eye-level → latest screenshot)
Activity log: ReAct thought → tool actions → response per turn
Hit tracking: OnActorHit plugin event → collision count, impulse, impact point
Floating draggable detail window per agent

📊 Statistics Panels

Embodied Agent Statistics: aggregate map (2D top-down, all agents), collision/speed time-series charts
Coding Agent Verifier: rule-based (scene collision count) + VLM-based (Claude reads screenshot, scores 1–10)
MetricsHub: server-side 5s sampling → real-time SVG LineChart / MultiLineChart

📁 Asset Content Drawer

Built once at startup from live UE Python scan (EditorAssetLibrary.list_assets)
UE Content Browser–style navigation: folder → shows subfolders + direct assets
Left sidebar: global category filter (Buildings / Trees / Vehicles / etc.)
Search: recursive subtree, category-aware
Click any asset → inserts correct spawn command into chat

🧪 Multi-Agent Testbed

Auto-start PIE (polls until confirmed active)
Spawn N agents at configurable radius, broadcast goal
Testbed log with per-step status

🔧 Developer Tools

Mock mode: replay pre-recorded sessions without GPU
Session management: slot-based with heartbeat and TTL
Skills: pre-made prompts for city layout, weather, navigation
Learned tools: Claude can save custom tool recipes

Note on demo assets: The featured demo showcases scenes built with high-quality commercial 3D assets (buildings, vehicles, characters, etc.) that are not included in the open-source release due to licensing restrictions. The redistributable Minimal build ships with a different set of freely licensed assets, so the visual appearance will differ from the demo. The functionality and workflow remain the same.

Quick Start (Google Colab — no local GPU needed)

Run all cells in order. Setup takes ~5 minutes. Requires a free Colab GPU runtime and an Anthropic API key.

Quick Start (Linux — One-Command Install)

For end users on a Linux machine with an NVIDIA GPU. For internal multi-user shared servers, see Linux — Shared Server below.

Prerequisites

OS: Linux (Ubuntu 20.04+ recommended)
GPU: NVIDIA GPU with 8GB+ VRAM (tested on L40S, T4, A100)
NVIDIA drivers: 525+ with Vulkan support
Node.js: 18+
Python: 3.9+
Disk: ~40 GB free (15 GB download + 21 GB extracted)

1. Download the Minimal SimWorld Binary

# Download (~15 GB compressed, ~21 GB extracted)
wget -O SimWorld-Studio-Minimal.tar.gz \
    https://huggingface.co/datasets/SimWorld-AI/SimWorld-Studio/resolve/main/SimWorld-Studio-Minimal.tar.gz

tar xzf SimWorld-Studio-Minimal.tar.gz

2. Install SimWorld Studio

pip install git+https://github.com/SimWorld-AI/SimWorld-Studio.git#subdirectory=packaging
npm install -g @anthropic-ai/claude-code

3. Authenticate with Claude

Option A — API Key:

export ANTHROPIC_API_KEY="sk-ant-..."

Get your key at console.anthropic.com.

Option B — Claude Code Login (no API key needed):

claude

This opens a browser for OAuth login. If on a headless server, use the API key option instead.

4. Launch (one command)

simworld-studio start

This will:

Detect your GPU and authenticate with Claude
Launch Unreal Engine (headless)
Wait for the engine to be ready
Start the Studio web server
Print the URL to open in your browser

On multi-GPU systems, it will ask which GPU to use (or pass --gpu INDEX). For remote servers, it auto-detects your IP and prints SSH tunnel instructions.

Options:

--gpu INDEX    GPU to use (auto-detected if omitted)
--port PORT    Web UI port (default: 3002)
--binary PATH  Path to SimWorld-Studio-Minimal directory

Try: "Set up the environment with a sunny sky, then build a small neighborhood with 4 houses and trees"

Quick Start (Windows — Local Dev)

Prerequisites

Unreal Engine 5.3 (Epic Games Launcher)
Node.js 18+
Claude Code CLI: npm install -g @anthropic-ai/claude-code

1. Configure paths

Edit SimWorld-Studio.ps1 lines 46–47:

$UeRoot    = "C:\Program Files\Epic Games\UE_5.3"
$UeProject = "E:\UE\SimWorld-copy\SimWorld.uproject"

2. Launch (one command)

.\SimWorld-Studio.ps1

The script will:

Kill stale port processes
Sync Cirrus player.js and build the React frontend
Start Cirrus signalling server (Pixel Streaming)
Launch UE Editor with MCP + Pixel Streaming flags
Wait for MCP port to be ready (up to 90s)
Start Web Server on port 3002
Stream all logs to terminal with [cirrus] / [server] prefixes

Open http://localhost:3002 in Chrome.

Options

.\SimWorld-Studio.ps1 -Port 3003 -McpPort 55560 -CirrusHttpPort 8687 -Gpu 1 -NoBuild

Quick Start (macOS — Web stack only, no UE)

Unreal Engine 5.3 has no native macOS build of this project, but the web stack (React UI, Node API, asset/skill/scene browsers, agent panel scaffolding) runs fine on Mac for frontend work and API exploration. UE-dependent features (viewport, scene spawn, agent control) will show ueConnected: false and are no-ops.

Prerequisites

Node.js 18+ (brew install node)

Run

./SimWorld-Studio-Mac.sh              # installs deps, builds frontend, starts backend on :3002
./SimWorld-Studio-Mac.sh --no-build   # reuse existing dist/
./SimWorld-Studio-Mac.sh --dev        # vite dev server on :5173 with HMR + backend on :3002

Then open http://localhost:3002 (or :5173 in --dev).

The script sets UNREAL_PORT=1 / UCV_PORT=1 so the backend's UE probes fail fast and stay quiet — no Cirrus, no UE Editor, no MCP subprocess required.

Quick Start (Linux — Shared Server)

See INTERNAL_SETUP.md for port assignment table and multi-user setup.

export UE_ROOT=/data/murray/ue/UE_5.3
export UE_PROJECT_PATH=/data/user/simworld_projects

simworld-studio start \
  --port 3004 --mcp-port 55561 \
  --cirrus-http-port 8687 --cirrus-ws-port 8688 --cirrus-sfu-port 8990 \
  --gpu 0

UE Plugin Extensions

The UnrealCV plugin has been extended with:

Command	Description
`vget /object/{name}/velocity`	Actor velocity [vx, vy, vz] cm/s
`vget /object/{name}/overlaps`	Currently overlapping actors (JSON)
`vget /object/{name}/nearby {r}`	Actors within radius r cm (JSON)
`vset /object/{name}/track_hits`	Bind `OnActorHit` — start recording physics collisions
`vget /object/{name}/hit_events`	Get & clear hit event queue (JSON)
`vget /camera/actor/{name}/lit`	Render from named actor's camera (FusionCamSensor or UCameraComponent)

Recompile required after updating plugin files (Ctrl+Alt+F7 in UE Editor).

Module READMEs

Module	README
Web Server (backend)	`web/server/README.md`
Frontend (React)	`web/src/README.md`
Web (full stack)	`web/README.md`
Gym / Navigation	`gym_env/README.md`
Co-Evolve	`co_evolve/README.md`

For Developers

Build from Source

# Requires access to the simworld_arena source repo
./build.sh

Release

./build.sh → creates dist/simworld_studio-{VERSION}.tar.gz
Upload to GitHub Releases
Update version.json

Troubleshooting

Symptom	Fix
`[cirrus] EADDRINUSE :8685`	`Get-NetTCPConnection -LocalPort 8685 \| Stop-Process -Force`
MCP port never ready	UE crashed — check `simworld_studio_workspace/logs/ue.log`
Pixel Streaming black screen	Wait 60s after UE launch; check `logs/cirrus.log` for `Streamer connected`
`Cannot generate unique name for X`	Fixed — actor names now include session suffix automatically
Agent not detected after PIE	Auto-discovery runs every 5s; or use ↻ Sync button in Agent panel
VLM Scoring fails	Ensure Claude Code is authenticated: `claude`
Camera tab black	Recompile UE plugin for `vget /camera/actor/{name}/lit`
`Vulkan memory crash` (Linux)	Use `--gpu 0` flag; install `vulkan-tools mesa-vulkan-drivers`
`MCP port not opening` (Linux)	Wait 60s more; check GPU drivers with `nvidia-smi`
`game module not found` (Linux)	Ensure you extracted the full archive; check `gym_citynav/Binaries/Linux/`
`CUDA context error` (Linux)	Set `--gpu INDEX` to isolate a single GPU
`Claude errors`	Run `claude login` or verify `ANTHROPIC_API_KEY` is set
`No GPU detected` (Linux)	Install NVIDIA drivers 525+; verify with `nvidia-smi`
Can't access UI remotely	Use SSH tunnel: `ssh -L 3002:localhost:3002 -L 8585:localhost:8585 user@server`

View Logs

# UE logs (Linux Minimal binary)
tail -f gym_citynav/Saved/Logs/gym_citynav.log

# Studio backend logs
tail -f simworld_studio_workspace/logs/server.log

License

Apache 2.0 — see LICENSE.

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

SimWorld Studio

Overview

Architecture

Component Map

Features

🏗️ Coding Agent (Scene Generation)

👁️ Live Viewport

🤖 Embodied Agent Panel

📊 Statistics Panels

📁 Asset Content Drawer

🧪 Multi-Agent Testbed

🔧 Developer Tools

Quick Start (Google Colab — no local GPU needed)

Quick Start (Linux — One-Command Install)

Prerequisites

1. Download the Minimal SimWorld Binary

2. Install SimWorld Studio

3. Authenticate with Claude

4. Launch (one command)

Quick Start (Windows — Local Dev)

Prerequisites

1. Configure paths

2. Launch (one command)

Options

Quick Start (macOS — Web stack only, no UE)

Prerequisites

Run

Quick Start (Linux — Shared Server)

UE Plugin Extensions

Module READMEs

For Developers

Build from Source

Release

Troubleshooting

View Logs

License

FilesExpand file tree

README.md

Latest commit

History

README.md

File metadata and controls

SimWorld Studio

Overview

Architecture

Component Map

Features

🏗️ Coding Agent (Scene Generation)

👁️ Live Viewport

🤖 Embodied Agent Panel

📊 Statistics Panels

📁 Asset Content Drawer

🧪 Multi-Agent Testbed

🔧 Developer Tools

Quick Start (Google Colab — no local GPU needed)

Quick Start (Linux — One-Command Install)

Prerequisites

1. Download the Minimal SimWorld Binary

2. Install SimWorld Studio

3. Authenticate with Claude

4. Launch (one command)

Quick Start (Windows — Local Dev)

Prerequisites

1. Configure paths

2. Launch (one command)

Options

Quick Start (macOS — Web stack only, no UE)

Prerequisites

Run

Quick Start (Linux — Shared Server)

UE Plugin Extensions

Module READMEs

For Developers

Build from Source

Release

Troubleshooting

View Logs

License