kubeshark

kubeshark

Kubeshark is an eBPF-based Kubernetes network observability tool that enables AI agents to query network data via the MCP interface.

MCPDevOpen sourcekubernetesnetwork-observabilityebpfmcp
Type
MCP
Transport
http
Open source
Yes
GitHub Stars
★ 12.0k
Source
mcp-github

Overview

Kubeshark uses eBPF to index cluster-wide network traffic at the kernel level, providing real-time query capabilities. It can decrypt TLS traffic without requiring key management or sidecars. AI agents can query network data through the MCP interface for troubleshooting and root cause analysis. Ideal for developers and operators who need to monitor and analyze Kubernetes cluster network traffic.

Capabilities

  • Index L4/L7 traffic
  • Decrypt TLS traffic
  • Query network data via MCP
  • Visualize network data
  • Download PCAP files

Use cases

Kubernetes network troubleshootingAPI call analysisRoot cause analysisNetwork traffic monitoring

Setup

Requires: Kubernetes 集群Helm
helm repo add kubeshark https://helm.kubeshark.com && helm install kubeshark kubeshark/kubeshark && kubectl port-forward svc/kubeshark-front 8899:80

This information was compiled by AI from public sources and may contain inaccuracies — please refer to the source.

FAQ

How do I connect an AI agent?

After installing with `brew install kubeshark`, run `claude mcp add kubeshark -- kubeshark mcp`.

What protocols are supported?

Supports multiple protocols including HTTP, gRPC, Redis, Kafka, DNS, and more.

Related skills