rcourtman/Pulse

What it solves

Pulse is a self-hosted infrastructure monitoring workspace that eliminates the need for a complex enterprise monitoring stack. It provides a unified view of live state, history, and alerts across diverse platforms like Proxmox, Docker, Kubernetes, TrueNAS, and various physical or virtual machines, helping operators find issues that dashboards alone cannot surface, such as failed backups or restart loops.

How it works

Pulse uses a combination of API-based monitoring (e.g., for Proxmox) and a unified agent installed on hosts to collect metrics, health data, and system state. It organizes this data into platform-specific views and cross-platform views for alerts and actions. The system includes "Pulse Patrol," which runs scheduled checks across infrastructure state and history, utilizing local or external AI models for analysis. It also provides an interactive Assistant and an MCP adapter for external AI clients like Claude Code.

Who it’s for

It is designed for system administrators and infrastructure operators who manage a mix of Proxmox, Kubernetes, Docker, and TrueNAS environments and want a centralized, operator-controlled monitoring solution.

Highlights

• Broad Platform Support: Covers Proxmox VE/PBS/PMG, Docker/Podman, Kubernetes, TrueNAS SCALE/CORE, VMware vSphere, and general Linux/Windows/macOS machines.
• AI-Powered Analysis: Pulse Patrol uses AI to perform scheduled health checks and identify problems requiring attention.
• Operator-Controlled Security: Features encrypted credentials at rest, scoped API tokens, and hardened agents with command execution disabled by default.
• Flexible Deployment: Supports Docker, Proxmox LXC, Linux, and Kubernetes/Helm installations.
• AI Integration: Includes an interactive Assistant and an MCP adapter for integration with external LLM clients.