Skip to main content
Glama

Blender Computer-Use – Isolierte virtuelle Maschine & MCP Server

Eine isolierte, GPU-partitionierte lokale Windows-11-Virtual-Machine-Umgebung, die über das Model Context Protocol (MCP) angebunden ist und speziell für KI-Codierungsagenten entwickelt wurde, um automatisierte Blender-UI-Tests, User-Story-Validierung und Computer-Use-Workflows durchzuführen, ohne den Host-Desktop zu stören.


1. Systemarchitektur

+---------------------------------------------------------------------------------------------+
|                          Coding Agents (Claude Code / OMP / Codex)                          |
+---------------------------------------------------------------------------------------------+
                                              │
                                              │ (MCP JSON-RPC via stdio)
                                              ▼
+─────────────────────────────────────────────────────────────────────────────────────────────+
|                         Host Layer: `blender-cu-vm-mcp` Server                             |
|  - MCP Tool Handler (Computer Use, Analytical Inspection, Blender Telemetry, File Staging)   |
|  - VM Lifecycle Manager (Hyper-V Socket / PowerShell WMI / Sub-2s Snapshot Reset)           |
+─────────────────────────────────────────────────────────────────────────────────────────────+
                                              │
                      ┌───────────────────────┴───────────────────────┐
                      │ Hyper-V Socket (HV-SOCK) / Internal VMSwitch  │
                      ▼                                               ▼
+───────────────────────────────────────────────+   +─────────────────────────────────────────+
|      Windows 11 Guest VM (Hyper-V)            |   |   Secondary (Fast CI): WSL2 Container   |
|  - NVIDIA GPU-PV (RTX 4080 Super vGPU: 8GB)   |   |   - Mesa / Direct3D 12 GPU Accel        |
|  - Virtual Display Driver (1080p Fixed 60Hz)  |   |   - Virtual X11 Display (Xvfb/Weston)   |
|  - Guest Agent Daemon (FastAPI / gRPC)        |   |   - Guest Agent Daemon (Linux)          |
|  - In-Process Blender Telemetry Bridge (bpy)  |   |   - Blender bpy IPC Bridge              |
+───────────────────────────────────────────────+   +─────────────────────────────────────────+

Related MCP server: kwin-mcp

2. Hauptfunktionen

  • Keinerlei Störung des Hosts: Synthetische Mausbewegungen, Ziehbewegungen, Klicks und Tastatureingaben werden ausschließlich innerhalb des Gastbetriebssystems ausgeführt. Ihr echter Cursor und Ihr Fensterfokus bleiben dabei vollständig unberührt.

  • Hardwarebeschleunigtes GPU-Rendering: NVIDIA GPU-PV verleiht der Gast-VM nahezu nativen Zugriff auf die RTX 4080 Super des Hosts (DirectX 12, Vulkan, OpenGL und CUDA/OptiX).

  • Deterministisches 1080p-Display: Der Open-Source-Virtual-Display-Treiber (IddSampleDriver) fixiert den virtuellen Bildschirm auf 1920x1080 bei 60 Hz mit 100 % (96 DPI) Skalierung und verhindert dadurch Koordinatendrift sowie einschlafende Monitore.

  • Hybrides multimodales Feedback: Die Agenten erhalten:

    1. Visuell: Framebuffer-Screenshots mit optionalen Koordinatengitter-Overlays.

    2. Analytisch: Windows-UI-Automation-Baum, Bounding Boxes und Fenster-Reaktionsfähigkeit.

    3. Tiefergehende Telemetrie: Blender-bpy-Zustand, aktive Modifier, Node-Tree-Verbindungen und Echtzeit-stdout/stderr-Protokolle.

  • State-Rollback in unter 2 Sekunden: Die schnelle Hyper-V-Snapshot-Wiederherstellung „setzt" die VM nach destruktiven oder experimentellen Läufen auf eine saubere Golden Base zurück.


3. Verzeichnisstruktur

blender-cu-vm/
├── host/                     # Host MCP Server & Hyper-V Controller
│   ├── mcp_server.py         # Stdio JSON-RPC MCP server with 14 tools
│   ├── hv_transport.py       # Hyper-V Socket (AF_HYPERV) & HTTP transport
│   ├── vm_controller.py      # PowerShell WMI lifecycle & snapshot manager
│   └── asset_sync.py         # Bi-directional file and addon staging
├── guest/                    # Guest Agent Daemon (runs inside VM)
│   ├── guest_daemon.py       # FastAPI HTTP/HV-SOCK unified server
│   ├── screen_capture.py     # DXGI Desktop Duplication & visual diffs
│   ├── input_controller.py   # Win32 SendInput (clicks, drags, typing)
│   ├── ui_automation.py      # Windows UI Automation tree inspector
│   └── video_recorder.py     # Hardware-accelerated NVENC MP4 recorder
├── blender/                  # Blender Embedded Runtime Bridge
│   ├── cu_telemetry_bridge.py # Non-blocking TCP telemetry server
│   ├── crash_interceptor.py  # C-level stdout/stderr stream tee
│   └── state_inspector.py    # Declarative scene invariant checker
├── scripts/                  # Automated Setup & Provisioning
│   ├── setup_vm_gpupv.ps1    # Automated Hyper-V Gen2 VM creator
│   ├── stage_gpupv_drivers.ps1 # NVIDIA GPU-PV driver packaging & injection
│   ├── setup_virtual_display.ps1 # Virtual display & autologon configuration
│   └── manage_golden_snapshot.ps1 # Instant snapshot creation & rollback
├── tests/                    # Verification & E2E Test Suite
│   ├── test_blender_user_story.py # 12-stage automated test suite
│   └── verify_isolation.py   # Zero host disturbance verification
├── mcp-config.json           # Registration snippet for Claude Code / OMP
└── README.md

4. Setup- und Installationsanleitung

Schritt 1: VM auf dem Host bereitstellen (PowerShell als Administrator)

cd C:\tmp\blender-cu-vm\scripts
.\setup_vm_gpupv.ps1 -VMName "Blender-CU-VM" -MemoryBytes 8GB -ProcessorCount 8

Schritt 2: NVIDIA-GPU-PV-Treiber vorbereiten

.\stage_gpupv_drivers.ps1 -VMName "Blender-CU-VM" -Mode "Stage"

Schritt 3: Gastbetriebssystem installieren und Umgebung einrichten

Innerhalb der Guest-VM (via PowerShell als Administrator):

# 1. Install GPU-PV drivers from staged directory
C:\Temp\NvidiaDrivers\install_gpupv_guest.bat

# 2. Configure Virtual Display & Auto-Logon
.\setup_virtual_display.ps1 -TargetWidth 1920 -TargetHeight 1080

# 3. Start Guest Daemon on boot
python C:\blender-cu-vm\guest\guest_daemon.py

Schritt 4: Golden-Base-Snapshot erstellen

.\manage_golden_snapshot.ps1 -VMName "Blender-CU-VM" -SnapshotName "golden_base" -Action "Create"

5. KI-Codierungsagenten über MCP verbinden

Fügen Sie Folgendes zu Ihrer ~/.claude.json or ~/.omp/agent/config.yml ein:

{
  "mcpServers": {
    "blender-cu-vm": {
      "command": "python",
      "args": [
        "C:\\tmp\\blender-cu-vm\\host\\mcp_server.py"
      ],
      "env": {
        "BLENDER_VM_NAME": "Blender-CU-VM",
        "BLENDER_GUEST_URL": "http://192.168.122.100:8000"
      }
    }
  }
}

6. Verifikationstests ausführen

Um alle Subsysteme zu verifizieren und die simulierte Blender-User-Story auszuführen:

python blender-cu-vm/tests/test_blender_user_story.py
python blender-cu-vm/tests/verify_isolation.py
F
license - not found
Not graded
quality - not tested
B
maintenance

Maintenance

Maintainers
Response time
0dRelease cycle
2Releases (12mo)
Commit activity

Resources

Unclaimed servers have limited discoverability.

Looking for Admin?

If you are the server author, to access and configure the admin panel.

Related MCP Servers

  • F
    license
    Not graded
    quality
    C
    maintenance
    Enables autonomous desktop automation by delegating tasks to vision-based agents operating within cloud-based virtual machine sandboxes. It allows users to manage VMs, execute complex computer tasks, and receive text-based screen summaries across Linux, Windows, and macOS environments.
    2
  • A
    license
    Not graded
    quality
    C
    maintenance
    Enables AI agents to automate Linux desktop GUI by launching and interacting with Wayland applications in isolated virtual KWin sessions, or connecting to live desktops for collaborative automation.
    39
    MIT

View all related MCP servers

Related MCP Connectors

  • Eyes and hands on real Windows PCs — observe, click, type via Glasswarp API.

  • Operate Linux, macOS and Windows from your LLM. Every action runs through an auditable allowlist.

  • Build, validate, and deploy multi-agent AI solutions from any AI environment.

View all MCP Connectors

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/trdillman/blender-computer-use-vm'

If you have feedback or need assistance with the MCP directory API, please join our Discord server