weibaohui/k8m

What it solves

K8M provides a lightweight, AI-powered dashboard for managing Kubernetes clusters. It simplifies complex cluster operations for developers and operators by replacing dense command-line interactions with an intuitive visual interface and AI-assisted troubleshooting.

How it works

Built with a Golang backend and a Baidu AMIS frontend, K8M acts as a control console that connects to Kubernetes clusters via the kom API client. It integrates Large Language Models (LLMs)—including built-in Qwen2.5-Coder and support for DeepSeek-R1 or private models via Ollama—to provide intelligent assistance. It also implements the Model Context Protocol (MCP) to allow LLMs to execute specific cluster operations through a set of built-in tools.

Who it’s for

It is designed for Kubernetes developers and DevOps/SRE engineers who need a simple, fast way to manage multiple clusters (including standard k8s, EKS, k3s, kind, and k0s) without relying solely on kubectl.

Highlights

• AI-Driven Operations: Offers AI-powered log diagnosis, YAML attribute translation, resource guides, and command recommendations.
• MCP Integration: Acts as an MCP Server with 49 built-in k8s tools, allowing LLMs to perform complex cluster operations with user-specific permissions.
• Multi-Cluster Management: Automatically scans kubeconfig files, supports heart-beat detection, and provides granular RBAC for users and groups.
• Pod Management: Includes a visual file explorer for Pods (edit/upload/download) and real-time log streaming with highlighted search.
• Cluster Health: Supports scheduled inspections with custom Lua rules and AI-summarized reports sent via webhooks (DingTalk, WeChat, Feishu).
• Deployment: Distributed as a single executable for easy installation across Linux, macOS, and Windows.