By name: PBnJ | PMOVES + Pinokio Bridge description: | Multi-target deployment bridge for PMOVES.AI infrastructure. Manages local Docker Compose, Kubernetes AI Lab, KVM4 production, VPS fleet deployment, homelab bare-metal, and edge (Jetson) nodes from a single Pinokio control panel. Built-in gepeto + pinokio skills. Covers: Z890, 5090, 4090, SPARK, R9700, JONS×3, KVM4-1/4-2, KVM2. keywords: deploy, infrastructure, docker, kubernetes, kvm4, vps, tailscale, glances, network, diagnostics, fleet, jetson, spark, rdna4, edge, gepeto, pinokio version: 1.1.0 category: Infrastructure/Deployment tier: 1 agent_class: Standard agent_id: pmoves_pbnj_bridge pinokio_skills: native: - pinokio # Built-in app discovery + launch - gepeto # Built-in code assistant (~/.agents/skills/gepeto) - pterm # Pinokio terminal interface pmoves: - pmoves-services # Docker Compose profile controls - pmoves-remote # Headscale/RustDesk remote access - pmoves-agent-zero - pmoves-model-registry - cipher # MoF/P7 state management
PBnJ | PMOVES + Pinokio Bridge
Agent Class: Standard (Pmoves-)
Category: Infrastructure/Deployment
Version: 1.1.0
Tier: 1 (Core Infrastructure)
Status: Active — homelab + VPS + edge fleet, gepeto + pinokio native skills
Capabilities
| Command | What It Does |
|---|---|
local-up |
Start local Docker Compose stack (Z890 dev) |
local-down |
Stop local Docker Compose stack |
local-logs |
Stream logs from local services |
lab-up |
Apply AI Lab manifests to Kubernetes |
lab-down |
Delete AI Lab from Kubernetes |
kvm4-up |
Deploy KVM4 production stack |
kvm4-down |
Stop KVM4 production stack |
kvm4-1-deploy |
Deploy API services to KVM4-1 (Agent Zero, TZ, Hi-RAG) |
kvm4-2-deploy |
Deploy data services to KVM4-2 (Supabase, Qdrant, Neo4j) |
kvm2-deploy |
Deploy exit proxy to KVM2 (Nginx) |
4090-deploy |
Deploy coding workstation to 4090 laptop |
4090-models |
Pull models to 4090 (~50GB) |
4090-status |
Check 4090 node status |
spark-deploy |
Bootstrap DGX Spark (GB10 Grace-Blackwell, 128GB) |
spark-models |
Pull models to SPARK (Gemma 4 31B FP16, Nemotron-Super-49B) |
spark-status |
Check SPARK node health (Tailscale + Ollama) |
rdna4-deploy |
Bootstrap RDNA4 workstation (dual R9700, ROCm 7.1) |
rdna4-status |
Check RDNA4 node health (llama-server + rocm-smi) |
jons-deploy |
Bootstrap all 3 Jetson Orin Nano Super units (JetPack 6.x) |
jons-status |
Check Jetson edge fleet health (mesh-agent + ollama) |
fleet-status |
Full fleet: all nodes Tailscale ping + service health |
status |
Kubernetes deployment status |
vps-status |
Tailscale ping + runner status for all VPS nodes |
Trigger Phrases (Pinokio 7 Interpreter)
Uses built-in gepeto and pinokio skills for natural language routing.
| Phrase | Action | Script |
|---|---|---|
"start local dev" |
Launch Docker Compose | local-up.json |
"stop local dev" |
Stop Docker Compose | local-down.json |
"show logs" |
Stream local service logs | local-logs.json |
"deploy to lab" |
Apply K8s manifests | lab-up.json |
"deploy to production" |
Apply KVM4 K8s stack | kvm4-up.json |
"deploy kvm4-1" |
API services to KVM4-1 | kvm4-1-deploy.json |
"deploy kvm4-2" |
Data services to KVM4-2 | kvm4-2-deploy.json |
"deploy kvm2" |
Exit proxy to KVM2 | kvm2-deploy.json |
"deploy 4090" |
Coding workstation to 4090 | 4090-deploy.json |
"deploy spark" |
Bootstrap DGX SPARK node | spark-deploy.json |
"deploy rdna4" |
Bootstrap RDNA4 workstation | rdna4-deploy.json |
"deploy jetsons" |
Bootstrap all 3 Jetson JONS | jons-deploy.json |
"check cluster status" |
K8s status | status.json |
"check vps fleet" |
All VPS node health | vps-status.json |
"check fleet" |
Full fleet health (all nodes) | fleet-status.json |
"stop everything" |
Stop all stacks | lab-down.json → kvm4-down.json → local-down.json |
Fleet Nodes
Homelab (Bare Metal)
| Node | Role | Deploy Script |
|---|---|---|
Z890 (pmoves-z890) |
Infrastructure coordinator, Docker, CI | local-up.json |
5090 (pmoves-5090) |
GPU compute, voice, model training | (via pmoves-services) |
SPARK (pmoves-gb10-spark) |
DGX Grace-Blackwell, 128GB unified, ARM64 inference | spark-deploy.json |
RDNA4 (pmoves-rdna4) |
Dual R9700 workstation, ROCm 7.1, 64GB VRAM | rdna4-deploy.json |
4090 Laptop (pmoves-4090) |
Field agent, mobile | 4090-deploy.json |
Edge Nodes (Jetson Orin Nano Super × 3)
| Node | Role | Deploy Script |
|---|---|---|
JONS-1 (pmoves-jons-1) |
Edge inference, Whisper, YOLOv8 | jons-deploy.json |
JONS-2 (pmoves-jons-2) |
Edge inference, Whisper, YOLOv8 | jons-deploy.json |
JONS-3 (pmoves-jons-3) |
Edge inference, Whisper, YOLOv8 | jons-deploy.json |
Jetson specs: 6-core Cortex-A78AE, Ampere SM_87 1024 CUDA cores, 8GB LPDDR5 (102 GB/s), 67 TOPS.
JetPack 6.x (L4T 36.x). Built-in Pinokio gepeto + pinokio skills.
VPS (Hostinger)
| Node | Role | Deploy Script |
|---|---|---|
| KVM4-1 | API gateway (TZ, A0, Hi-RAG, Archon) | kvm4-1-deploy.json |
| KVM4-2 | Data/storage (Supabase, Qdrant, Neo4j) | kvm4-2-deploy.json |
| KVM2 | Exit proxy (Nginx) | kvm2-deploy.json |
Network Tools (Optional)
When network tools are installed (app/env exists):
- Glances — System monitor (venv or Docker mode)
- Network Diagnostics — Connectivity checks
- DNS Flush — Clear DNS cache
- Winsock Reset — Reset network stack
- Firewall Rules — Add PMOVES service rules
- Docker Network Cleanup — Prune stale networks
Prerequisites
- Docker Desktop with Compose v2+
- Tailscale client (for VPS deployments)
- kubectl (for Kubernetes targets)
pmoves/env.sharedbootstrapped