OpenSourceProjects logo
homelab-monitor logo

homelab-monitorSelf-hosted homelab dashboard in one container — GPU, local-AI VRAM, Docker, systemd, energy costs. Read-only MCP server so AI agents can explore it.

Self-hosted homelab dashboard in one container — GPU, local-AI VRAM, Docker, systemd, energy costs. Read-only MCP server so AI agents can explore it.

234 stars
37 forks
Python
MIT
homelab-monitor screenshot

homelab-monitor

A self-hosted dashboard that consolidates monitoring across all your homelab machines, GPUs, and AI models in a single container. Includes an AI agent powered by local models that answers questions about your infrastructure and can perform approved actions without cloud dependencies.

Key Features

  • Unified Dashboard: Monitor GPU usage, VRAM, Docker containers, systemd services, and energy costs across multiple machines from one interface
  • AI Agent Interface: Ask natural language questions about your homelab; the agent analyzes data and provides actionable insights with required approvals
  • MCP Server: Exposes homelab data as a read-only Model Context Protocol server, enabling AI agents to explore and understand your infrastructure
  • Local Model Support: Integrates with Ollama, vLLM, llama.cpp, and other local model servers on your network
  • Policy-Based Automation: Run routines on schedules with configurable access levels per machine and a fixed catalogue of safe actions
  • Multi-Machine SSH Access: Monitors Linux, Raspberry Pi, and Windows machines over SSH without requiring agents on each
  • Activity Logging: Complete audit trail of what was read, checked, and executed with policy decisions

Use Cases

  • Infrastructure Management: Monitor distributed homelab resources including GPU health, container status, and disk space across all machines
  • Cost Tracking: Measure and analyze energy consumption costs for running local AI models and services
  • Proactive Maintenance: Set up automated routines that detect issues like disk space problems or failed services and propose fixes
  • AI Agent Exploration: Provide AI agents with safe, read-only access to understand your infrastructure topology and resource allocation

Who Is It For

Homelab enthusiasts, self-hosted AI practitioners, and sysadmins running distributed infrastructure with GPUs and local models. Ideal for those who want comprehensive monitoring and AI-powered insights without relying on cloud services or installing agents on every machine.

Trending Open Source Projects