Kmcp
CLI tool and Kubernetes Controller for building, testing and deploying MCP servers
What is Kmcp?
Kmcp is a Model Context Protocol (MCP) server that allows AI assistants like Claude, Cursor, and VS Code to cli tool and kubernetes controller for building, testing and deploying mcp servers
CLI tool and Kubernetes Controller for building, testing and deploying MCP servers
This server falls under the Cloud Services category on MCPgee, the world's largest MCP server directory with 33,000+ servers.
Features
- CLI tool and Kubernetes Controller for building, testing and
Use Cases
Maintainer
Works with
Installation
Manual Installation
npx kmcpConfiguration
Configuration Details
claude_desktop_config.json
Performance
Response Metrics
Resource Usage
How to Set Up and Use Kmcp
kMCP is a CLI tool and Kubernetes Controller that streamlines the full lifecycle of MCP servers — from rapid scaffolding with FastMCP (Python) or MCP Go SDK, to packaging as containers, to production deployment on Kubernetes via Custom Resource Definitions. It provides a Transport Adapter layer that routes external traffic over HTTP, WebSocket, and Server-Sent Events, and integrates with Kubernetes secrets for secure credential management. Developers use kMCP to go from a local MCP prototype to a cloud-native, scalable service without writing Kubernetes manifests by hand.
Prerequisites
- A Unix-like shell (bash/zsh) or Windows with WSL for the install script
- A running Kubernetes cluster (local like kind/minikube or cloud-managed) for production deployments
- Python 3.10+ with FastMCP, or Go 1.21+ with the MCP Go SDK, depending on the language you choose
- kubectl configured and pointing at your target cluster
- An MCP-compatible client such as Claude Desktop or Claude Code
Install the kmcp CLI
Run the official install script which downloads the correct binary for your platform and places it on your PATH.
curl -fsSL https://raw.githubusercontent.com/kagent-dev/kmcp/refs/heads/main/scripts/get-kmcp.sh | bashVerify the installation
Confirm kmcp is available and check the available subcommands.
kmcp --helpScaffold a new MCP server project
Use the init command to create a project skeleton. Choose FastMCP for Python or the MCP Go SDK based on your preferred language.
kmcp init my-server --lang python
cd my-serverRun the server locally for testing
Start the server in local mode to iterate quickly before deploying. kmcp handles transport setup so your MCP client can connect immediately.
kmcp runBuild a container image
Package the server as a Docker image ready for Kubernetes. kmcp generates the Dockerfile and builds it in one command.
kmcp build --tag my-registry/my-server:latestDeploy to Kubernetes
Push the image and apply the generated Kubernetes manifests including CRDs and the Transport Adapter. kmcp installs the controller and creates the MCPServer custom resource.
kmcp deploy --image my-registry/my-server:latest --namespace mcp-serversConfigure your MCP client
Point Claude Desktop or another MCP client at the HTTP endpoint exposed by the Transport Adapter. Replace the URL with your cluster's external address.
Kmcp Examples
Client configuration
Configure Claude Desktop to connect to a locally running kMCP server via stdio using npx.
{
"mcpServers": {
"kmcp": {
"command": "npx",
"args": ["kmcp", "run", "--stdio"]
}
}
}Prompts to try
Example prompts once your kMCP-managed server is connected to an MCP client.
- "Scaffold a new FastMCP Python server called weather-api"
- "Build a container image for my MCP server and tag it for my registry"
- "Deploy my MCP server to the mcp-servers namespace in my Kubernetes cluster"
- "Show me the status of all MCPServer resources in the cluster"Troubleshooting Kmcp
kmcp command not found after running the install script
The script installs the binary to ~/bin or /usr/local/bin. Run `export PATH=$PATH:~/bin` or restart your shell, then verify with `which kmcp`.
kmcp deploy fails with 'no CRD found' error
The kMCP Kubernetes controller must be installed first. Run `kmcp install-controller` to apply the CRDs and controller deployment to your cluster before deploying servers.
MCP client cannot connect to the Transport Adapter endpoint
Verify the service is exposed with `kubectl get svc -n mcp-servers`. If using a LoadBalancer type, wait for an external IP to be assigned. For local clusters, use `kubectl port-forward` to expose the port.
Frequently Asked Questions about Kmcp
What is Kmcp?
Kmcp is a Model Context Protocol (MCP) server that cli tool and kubernetes controller for building, testing and deploying mcp servers It connects AI assistants to external tools and data sources through a standardized interface.
How do I install Kmcp?
Follow the installation instructions on the Kmcp GitHub repository. Clone the repo, install dependencies, and add the server config to your AI client.
Which AI clients work with Kmcp?
Kmcp works with all major MCP-compatible AI clients including Claude Desktop, Claude Code, Cursor, VS Code (GitHub Copilot), Windsurf, and Cline.
Is Kmcp free to use?
Yes, Kmcp is open source and available under the Apache-2.0 license. You can use it freely in both personal and commercial projects.
Kmcp Alternatives — Similar Cloud Services Servers
Looking for alternatives to Kmcp? Here are other popular cloud services servers you can use with Claude, Cursor, and VS Code.
Open WebUI
★ 138.2kUser-friendly AI Interface (Supports Ollama, OpenAI API, ...)
Anything LLM
★ 60.4kThe all-in-one AI productivity accelerator. On device and privacy first with no annoying setup or configuration.
LocalAI
★ 46.4kLocalAI is the open-source AI engine. Run any model - LLMs, vision, voice, image, video - on any hardware. No GPU required.
Nacos
★ 33.0kan easy-to-use dynamic service discovery, configuration and service management platform for building AI cloud native applications.
Xiaozhi ESP32
★ 26.7k本项目为xiaozhi-esp32提供后端服务，帮助您快速搭建ESP32设备控制服务器。Backend service for xiaozhi-esp32, helps you quickly build an ESP32 device control server.
Gateway
★ 11.8kA blazing fast AI Gateway with integrated guardrails. Route to 1,600+ LLMs, 50+ AI Guardrails with 1 fast & friendly API.
Browse More Cloud Services MCP Servers
Explore all cloud services servers available in the MCPgee directory. Each server includes setup guides for Claude, Cursor, and VS Code.
Set Up Kmcp in Your Editor
Choose your AI client for step-by-step setup instructions.
Quick Config Preview
Add this to your claude_desktop_config.json or .cursor/mcp.json
Ready to use Kmcp?
Browse our complete directory of 33,000+ MCP servers, read setup guides for your editor, and start building with the Model Context Protocol.