See every container, VM, and service running across a home lab in one dashboard.
Get alerted when a service goes unhealthy, a host stops reporting, or a container image falls behind.
Check host resource usage like CPU, memory, and disk across Docker, Proxmox, and Kubernetes hosts.
| techdox/trove | open-gitagent/langship.sh | tryharder-l/ninimenu | |
|---|---|---|---|
| Stars | 48 | 48 | 48 |
| Language | Go | Go | Go |
| Setup difficulty | moderate | hard | moderate |
| Complexity | 3/5 | 4/5 | 3/5 |
| Audience | ops devops | ops devops | general |
Figures from each repo's GitHub metadata at analysis time.
Requires Docker Compose per host, plus a separate agent deployment for Kubernetes or bare-metal Linux, dashboard has no login by default.
Trove is a self-hosted dashboard that automatically builds an inventory of everything running across a home lab or small infrastructure setup. It works by running small agent programs next to your actual workloads, whether they are Docker containers, a Kubernetes cluster, Proxmox virtual machines, or plain Linux servers, and each agent reports what it sees back to one central Trove server. The project's one hard rule is that it is read only. Agents only ever list and read information from the platforms they watch, and there is no way, by design, for Trove to restart, deploy, exec into, or otherwise change anything it monitors. It is meant to be the place you start looking when something seems off, not a tool for managing or controlling your systems. Once running, the dashboard shows a combined catalogue of containers, Kubernetes workloads with their pods grouped under each deployment, Proxmox virtual machines and containers, and Linux system services, all organized by host. It tracks whether each service is healthy, whether an agent has gone quiet and stopped reporting, host level resource usage like CPU, memory, and disk, and whether a running container's image has fallen behind the latest tagged version. It can also send alerts through webhooks, Discord, or ntfy when something goes unhealthy, disappears, or falls behind, and can send a scheduled summary email. Setting it up means running the Trove server plus one agent for each platform you want to watch, most easily through Docker Compose using example configuration files provided for Docker hosts and Proxmox. Kubernetes and bare metal Linux agents are deployed separately as a cluster deployment or a systemd service. Each agent needs its own token to authenticate with the server, and by default the dashboard itself has no login, so the project recommends keeping it on a private network or placing it behind an authenticating proxy. This is aimed at people running home labs or small self-hosted infrastructure who want one place to see what is actually running without granting a tool any power to change it.
A read-only dashboard that auto-discovers everything running across Docker, Kubernetes, Proxmox, and Linux hosts in a home lab, without any ability to change them.
Mainly Go. The stack also includes Go, Docker, Kubernetes.
Free to use, modify, and distribute for any purpose, including commercial use, as long as you keep the copyright and license notices.
Setup difficulty is rated moderate, with roughly 30min to a first successful run.
Mainly ops devops.
This repo across BitVibe Labs
double-check against the repo, no cap.