K8m, developed by Weibaohui, is an AI-driven Mini Kubernetes AI Dashboard lightweight console tool for the mcp platform. This MCP server is primarily built with Golang for its backend and Baidu AMIS for its frontend. K8m provides a unified interface for connecting to diverse Kubernetes clusters and various AI models. It offers capabilities for managing and analyzing Kubernetes environments, simplifying complex operations for developers, operators, and even beginners. Its design focuses on intelligent assistance for cluster management.

What does K8m connect to?

K8m connects to Kubernetes clusters, including standard k8s, AWS EKS, k3s, kind, and k0s, along with other cluster types. It also integrates with various AI models. K8m uses "kom" as its Kubernetes API client. Built-in AI models include Qwen2.5-Coder-7B and DeepSeek-R1-Distill-Qwen-7B. Additionally, K8m supports external private AI models, notably Ollama, expanding its AI capabilities. Its core function is as an MCP server.

Core features and tools

K8m exposes 49 built-in Kubernetes multi-cluster MCP tools, enabling over a hundred cluster operations. Derived from its "kom" library, these tools include creation, update, deletion, and retrieval of Kubernetes resources. It supports SQL-like querying of k8s resources and POD internal file operations (upload, download, delete). High-frequency actions like deployment restart, scaling, and start/stop are available. AI-driven functions provide K8s resource intelligent analysis, event consultation, log analysis, and Cron expression parsing.

Who is it for?

K8m targets developers and operators who manage Kubernetes environments. It is also suitable for beginners learning Kubernetes management due to its AI-assisted functions. This tool aids in simplifying complex tasks through natural language commands and provides intelligent analysis. Skip this if your workflow does not involve Kubernetes cluster management or AI model integration.

How to get started

The authentication model involves a default username and password ("k8m" for both) upon initial setup, which users are advised to change. For MCP calls, a JWT token is automatically added to the Authorization header, using stdio and sse transport protocols.

Use cases

• AI-assisted Kubernetes cluster management: Simplify tasks like checking pod status, scaling deployments, or analyzing ingress failures through natural language commands.
• Intelligent analysis and troubleshooting: Perform operations such as analyzing resource bottlenecks across clusters, generating deployment recommendations, and correlating pod failures with configuration changes.
• Multi-cluster operations: Manage and perform operations across multiple Kubernetes clusters, with automatic cluster recognition and registration.

K8m provides an AI-driven dashboard for managing various Kubernetes clusters and integrating with multiple AI models. Its Golang backend and Baidu AMIS frontend deliver a lightweight console for developers, operators, and beginners. It offers many tools for resource management and intelligent analysis, enhancing operational efficiency.