Infrastructure

Computer Node.
Cloud + Home Server Bridge.

A persistent WebSocket relay that connects your VP3 cloud account to your home server. Local AI inference, brain sync, video rendering, and task scheduling — all on your hardware, all in real time.

VP3 CLOUD Social + Commerce + AI api/vp3/ • MySQL • OB1 WSS :3077 YOUR HOME NODE Cortex v1.7.1 — Node.js server.js Brain Render Ollama Sched Whisper Claude Plugins CONNECTING... Request Response Render Job
How the Connection Works

Your VP3 Home Server (Node.js) establishes a persistent WebSocket connection to the VP3 cloud on port 3077. Once authenticated, it becomes a two-way relay for every subsystem.

1
Home Server Boots
2
WSS Handshake
3
Token Auth
4
Brain Sync
5
Modules Online
6
Connected
Zero Cloud Dependency
If your internet drops, the Home Server continues running locally. Brain entries, AI queries via Ollama, and scheduled tasks all work offline. When connectivity returns, the WebSocket reconnects and syncs the delta automatically.
What Runs on Your Home Node

Six core modules run on your home server. Each one communicates with the VP3 cloud through the WebSocket relay but can operate independently when disconnected.

🧠
Brain Sync

Your Computer Brain entries are mirrored between cloud and home server. New entries created offline are queued and pushed when the connection resumes. Conflict resolution uses last-write-wins with full audit trail.

BidirectionalOffline-Safe
🦙
Ollama LLM

Local AI inference through Ollama. Llama, Mistral, Gemma, Phi, or any open model. OB1 routes queries to Ollama first (free, zero latency) and falls back to Claude API only when needed. The source badge in chat shows which engine answered.

Local-FirstFree Inference
🎬
Remotion Renderer

Server-side video rendering powered by Remotion. Create product videos, social clips, and audio visualizers without touching a video editor. Jobs queue through the home server and render locally — no cloud GPU costs.

GPU-LocalQueued
📅
Task Scheduler

Cron-style task scheduling for automations, brain backups, sub-brain polling, and data mining. Tasks run on your hardware on your schedule. Results sync back to your cloud account through the WebSocket relay.

CronAutomated
🎙
Whisper Transcription

Local speech-to-text via OpenAI Whisper running on your hardware. Voice notes, recordings, and audio messages are transcribed locally — no audio data leaves your machine. Results feed into your brain entries.

Local STTPrivate
🔌
Plugin System

Extend your home node with custom plugins. The CSF validator scans all plugin scripts for dangerous patterns before execution. Plugins can register their own scheduled tasks, brain hooks, and WebSocket message handlers.

ExtensibleCSF-Validated
What Moves Between Cloud & Home

The WebSocket relay carries structured messages, not raw data. Every packet has a type, a target module, and a payload. Nothing syncs unless you've toggled the corresponding system on in the Data Control Center.

Cloud → Home

Commands and data that flow from your VP3 cloud account down to your home server.

  • AI query routing (when Ollama is preferred over Claude)
  • Brain entry sync — new entries created in-app push to home
  • Video render jobs queued from the web interface
  • Scheduled task definitions and cron updates
  • Plugin install and configuration commands
  • Sub-brain polling targets and data mining rules

Home → Cloud

Results and state that flow from your home server back up to your VP3 account.

  • Ollama AI responses routed back to OB1 chat
  • Brain entries created offline — queued delta sync
  • Rendered video files uploaded to your account storage
  • Scheduled task results and completion status
  • Whisper transcription results from local audio
  • Home server health, uptime, and resource metrics
Data Control Center Integration
Every subsystem that syncs through the Computer Node is controlled by the Data Control Center toggles. If you turn off Brain Sync, no brain data crosses the wire. If you turn off AI Services, Ollama still works locally but doesn't report back to cloud. The connection is always available — what travels over it is entirely up to you.
Requirements & Installation

The VP3 Home Server runs on any machine with Node.js. No special hardware required — a Raspberry Pi works, a laptop works, or deploy it on the VP3 Cauldron for the full experience.

System Requirements

Minimum specs to run the VP3 Home Server and establish the Computer Node connection.

  • Node.js 18 or higher
  • 2GB RAM minimum (4GB+ recommended for Ollama)
  • Port 8420 available (configurable in config)
  • Internet connection for initial auth (optional after)
  • Ollama installed for local AI (optional but recommended)
  • Remotion CLI + Renderer for video jobs (optional)

Quick Start

Four commands to get your home node connected to your VP3 account.

$ cd vp3-home
$ npm install
$ npm install @remotion/cli @remotion/renderer
$ node server.js

The server auto-detects Ollama, checks for Remotion, and establishes the WebSocket connection to your VP3 account. Status is visible in the OB1 chat header.

Connection Security

The Computer Node connection is secured at every layer. No data travels in plaintext, and no module can bypass the authentication handshake.

🔒
WSS Encryption

All traffic between cloud and home runs over WebSocket Secure (WSS). TLS 1.3 with certificate pinning. No fallback to unencrypted WebSocket connections.

🔑
Token Authentication

The home server authenticates with your VP3 HMAC token on every connection. Tokens rotate automatically. Stolen tokens are invalidated server-side within 60 seconds of a new connection.

🛡
CSF Validation

Every plugin script and scheduled task passes through the CSF validator before execution. 20 dangerous patterns are blocked. The energy grid auto-quarantines nodes with more than 15 errors per hour.

💾
Local Encryption

Brain data stored on the home server is encrypted at rest with hardware-backed keys (when running on the Cauldron) or AES-256 file encryption on standard hardware.

VP3 Computer Node vs. Traditional Cloud

Most platforms force you to choose: cloud convenience or local privacy. VP3 gives you both.

Capability VP3 Computer Node Traditional Cloud-Only
AI InferenceLocal (Ollama) + Cloud fallbackCloud only, per-token billing
Data LocationYour hardware + selective cloud syncVendor servers only
Offline ModeFull functionality, delta sync on reconnectNo offline support
Video RenderingLocal GPU, no cloud costsCloud GPU, pay per minute
Privacy ControlPer-system toggles, zero telemetryAll-or-nothing, opaque telemetry
LatencySub-10ms for local AI queries200-500ms round trip
CostFree for local operationsPer-request or subscription
What You Can Build With This

The Computer Node isn't just infrastructure — it unlocks capabilities that are impossible with cloud-only platforms.

🏠
Family Home Server

Run a single Cauldron or Raspberry Pi as the family hub. Every family member's VP3 account connects to the same home node. Shared brain, shared AI, shared media library — all local.

Family
🏪
Merchant Automation

Schedule product updates, inventory syncs, and order fulfillment tasks on your home node. Render product videos automatically. Run A/B price testing through scheduled brain analysis.

Commerce
🎵
Audio Production Pipeline

Transcode, master, and distribute .VP3 audio files through your home server. Render album art videos with Remotion. Auto-tag with ISRC metadata. Queue releases on a schedule.

Media
🧪
AI Research Node

Run multiple Ollama models locally and compare outputs. Feed results into your PLLM. Build specialized knowledge bases from local data processing without sending anything to the cloud.

Research
📊
Sub-Brain Data Mining

The scheduler polls external data sources on your behalf. RSS feeds, price trackers, social signals — all collected locally, filtered through your brain rules, and surfaced to OB1 as actionable insights.

MiningScheduled
🛡
Off-Grid Operation

Pair the Cauldron with the Solar Frame and run your entire VP3 instance off-grid. Local AI, local storage, local rendering. Connect to the internet when you want, not because you have to.

SolarSelf-Sufficient
Contribute Power. Mine VP3 Coin.

Your Computer Node doesn't just serve you — it can power the entire VP3 platform. Opt in to process AI queries, render videos, and index brains for other users. Earn VP3 Coin for every job your hardware completes.

Personal

Default mode. Your Cauldron runs your own VP3 instance. No network jobs, no mining.

Contributor

Process network jobs during idle time. Earn VP3 Coin passively. Your workloads always come first.

Miner / Investor

Dedicated node for the VP3 datacenter. 2x coin rate, early adopter multiplier, governance weight.

Learn About VP3 Network
6
Modules
<10ms
Local AI
0
Cloud Cost
24/7
Uptime

Your server. Your AI. Your data.

Set up your Computer Node in under a minute.

Create Your Account