StacklokLabs/mkp

Overview

MKP is a Model Context Protocol (MCP) server for Kubernetes. It exposes Kubernetes resources and operations through the MCP interface, enabling AI agents to interact with clusters. Built in Go, it provides a standardized way for LLMs to query and manage Kubernetes objects.

Best for

Best for
Developers building AI agents that need programmatic access to Kubernetes clusters

Use cases

• Query cluster state and resource configurations via natural language
• Automate routine Kubernetes operations through AI agent integrations
• Enable AI-assisted debugging and troubleshooting of Kubernetes workloads

How to use

Tools exposed

• get_resource
• list_resources
• apply_resource

Notes

MKP is a Model Context Protocol (MCP) server for Kubernetes. It exposes Kubernetes resources and operations through the MCP interface, enabling AI agents to interact with clusters. Built in Go, it provides a standardized way for LLMs to query and manage Kubernetes objects.

57 stars on GitHub. Last updated 2026-05-29. Licensed Apache-2.0.

Use cases

• Query cluster state and resource configurations via natural language
• Automate routine Kubernetes operations through AI agent integrations
• Enable AI-assisted debugging and troubleshooting of Kubernetes workloads

Pros

• Leverages the MCP standard for broad AI tool compatibility
• Lightweight Go implementation with minimal dependencies
• Directly addresses Kubernetes operational tasks without custom scripting

Cons

• Small community and limited adoption (57 stars)
• Requires MCP-compatible AI clients to be useful
• May lack advanced Kubernetes features or edge-case handling

Indexed from awesome-mcp-servers-punkpeye and enriched against its public facts.