Harvester HCI

Harvester MCP Server enables AI assistants like Claude Desktop and Cursor to interact with Harvester HCI clusters through a standardized protocol. Built in Go using the mark3labs/mcp-go SDK, it provides tools for managing both standard Kubernetes resources (pods, deployments, services) and Harvester-specific resources (virtual machines, images, volumes, networks). The server automatically connects to Kubernetes clusters using standard configuration paths and formats output in human-readable formats, making it ideal for DevOps teams who need AI assistance with Harvester cluster management and monitoring.