Real-time topology, live telemetry, and a local AI that actually knows your cluster — all on your own hardware.
⬇ Download · Features · Install · Screenshots · Development
InfraLens connects to the Proxmox API, discovers every node, VM, and LXC container in your cluster, and draws them as a live topology graph with streaming telemetry. A local Ollama model receives structured, real-time cluster context — so you can ask about your infrastructure in plain language instead of digging through logs and dashboards.
Everything stays local. Telemetry and chat run entirely on your network. No cluster data ever leaves your machine.
| Automatic discovery | Enumerates hosts, VMs, and LXC containers through the Proxmox API and lays them out as an interactive graph. |
| Live telemetry | Per-node CPU, RAM, disk, uptime, and network throughput — updated in real time. |
| Historical trends | CPU and memory sparklines per node, with hour / day / week views from Proxmox RRD data. |
| Power controls | Start, stop, reboot, and shut down VMs and containers directly from the app. |
| Cluster overview | Totals, aggregate resource meters, an attention list for offline or high-load nodes, and a recent-activity feed. |
| Service detection | Probes containers over SSH to identify Docker, databases, web servers, and more. |
| LLM diagnostics | Streams live cluster state into a local Ollama model for conversational troubleshooting. |
| Local only | No third parties. No telemetry. Your credentials never leave your machine. |
InfraLens ships as a native desktop app for macOS and Linux, built with Tauri 2. Grab the latest build from the Releases page.
| Platform | File |
|---|---|
| macOS · Apple Silicon | InfraLens_x.x.x_aarch64.dmg |
| macOS · Intel | InfraLens_x.x.x_x64.dmg |
| Linux · AppImage | InfraLens_x.x.x_amd64.AppImage |
| Linux · Debian / Ubuntu | InfraLens_x.x.x_amd64.deb |
🍎 macOS — first launch
The app is ad-hoc signed (not yet Apple-notarized), so macOS blocks it on first open. After dragging InfraLens into Applications, run this once:
xattr -cr /Applications/InfraLens.appThat clears the quarantine flag macOS adds to downloaded apps. Open InfraLens normally, then enable it under System Settings → Privacy & Security → Local Network so it can reach your Proxmox host.
You only do this once per install — after that, InfraLens opens straight to your topology.
🐧 Linux
AppImage — make it executable and run:
chmod +x InfraLens_x.x.x_amd64.AppImage
./InfraLens_x.x.x_amd64.AppImageDebian / Ubuntu — install the .deb:
sudo dpkg -i InfraLens_x.x.x_amd64.deb🧹 Uninstall & wipe data
sudo rm -rf /Applications/InfraLens.app
rm -rf ~/Library/Application\ Support/InfraLens
tccutil reset LocalNetworkA short wizard walks you through three steps:
- Proxmox — host / IP, API token user, token name, and value
- SSH probe (optional) — credentials to detect services inside LXC containers
- Local AI — pick an installed Ollama model for diagnostics
Credentials are saved locally to ~/Library/Application Support/InfraLens/config.json (macOS) or ~/.config/InfraLens/config.json (Linux) — and never leave your machine.
LLM diagnostics — query service status and trace bottlenecks in plain language with a local model.
Run from source or contribute.
Prerequisites — Node.js 18+, Python 3.10+, Rust (rustup), a Proxmox VE server with an API token, and Ollama running locally.
# 1. Clone
git clone https://github.com/AdemYahmadi/InfraLens.git
cd InfraLens
# 2. Backend -> http://127.0.0.1:8756
cd backend
python -m venv venv
source venv/bin/activate # Windows: venv\Scripts\activate
pip install -r requirements.txt
python main.py
# 3. Frontend (new terminal) -> http://localhost:5173
cd frontend
npm install
npm run devThe dev server proxies /api to the backend automatically — no .env needed.
InfraLens/
├── backend/ FastAPI · Proxmox discovery, SSH probe, Ollama chat
├── frontend/ React + Vite · topology UI, telemetry panels, chat
└── src-tauri/ Tauri 2 · bundles the backend, manages the window
Built for homelabbers. Released under the MIT License.