skyhook-io/radar

The missing open-source Kubernetes UI with a built-in MCP server for AI agents. See what's broken, why, and what changed. Issues, Topology, event timeline, Helm, GitOps, live service traffic, and cluster audits - all in one Go binary.

What it solves

Radar is an open-source Kubernetes UI designed to provide a comprehensive, real-time view of cluster state without requiring complex installation or cluster-side agents. It solves the problem of managing large-scale Kubernetes clusters by offering a fast, private, and airgapped-friendly dashboard that can run locally on a developer's laptop or as a shared in-cluster deployment.

How it works

Radar operates as a single binary that communicates directly with the Kubernetes API. It uses informers to watch the cluster in real-time and pushes updates to the browser via Server-Sent Events (SSE). It can be run as a local CLI tool (kubectl radar) or deployed via Helm for team access. For AI integration, it includes a built-in Model Context Protocol (MCP) server that allows AI agents to inspect and operate the cluster.

Who it’s for

It is built for Kubernetes administrators, DevOps engineers, and developers who need a high-performance UI for cluster visualization, resource management, and GitOps monitoring across various environments (GKE, EKS, AKS, minikube, etc.).

Highlights

  • Zero-Install Local Mode: Runs on your laptop and talks directly to the K8s API, keeping data private on your machine.
  • Topology Visualization: Interactive graphs showing resource relationships and network traffic flow.
  • Image Filesystem Viewer: Browse container image contents directly from the Pod view without needing to exec into containers.
  • GitOps Workspace: Dedicated monitoring and diagnosis for FluxCD and ArgoCD resources, including drift detection.
  • Resource Comparison: Side-by-side YAML diffs of any two resources of the same kind with noise-stripping for cleaner signals.
  • AI-Ready: Integrated MCP server for AI agent cluster diagnosis and operation.

Related

  • Project
    kubewall/kubewall

    A single-binary Kubernetes dashboard with multi-cluster management and AI integration for automated troubleshooting and resource monitoring.

  • Project
    weibaohui/k8m

    An AI-powered lightweight Kubernetes dashboard that simplifies cluster management through a visual interface, AI-assisted troubleshooting, and MCP-based tool execution.

  • Project
    kite-org/kite

    Kite is an open-source Kubernetes workspace for multi-cluster operations, observability, and AI-assisted troubleshooting.

  • Project
    Flux159/mcp-server-kubernetes

    An MCP server that enables AI assistants to manage Kubernetes clusters, perform diagnostics, and handle Helm operations using a standardized interface.

  • Project
    containers/kubernetes-mcp-server

    The Kubernetes MCP Server is a Go-based Model Context Protocol server that lets AI assistants and agents directly inspect and manage Kubernetes and OpenShift clusters through the Kubernetes API — not by wrapping kubectl or helm. It exposes tools for pods, namespaces, events, Helm releases, Tekton pipelines, KubeVirt VMs, and more, and can work across multiple clusters with no external command-line dependencies.