MCP ExplorerExplorer

Talk2 K8s

@chamodshehankaon 9 months ago
4 MIT
FreeCommunity
AI Systems
Talk to K8s Cluster with Kubernetes AI and MCP

Overview

What is Talk2 K8s

talk2-k8s is an AI-powered interface that allows users to interact with Kubernetes clusters using natural language, leveraging the capabilities of Large Language Models (LLM) through a Kubernetes MCP Server.

Use cases

Use cases for talk2-k8s include simplifying Kubernetes operations for developers, automating deployment processes, and enhancing user experience in managing cloud resources.

How to use

To use talk2-k8s, set up a Kubernetes cluster on Google Cloud, install the necessary tools like GCloud CLI and kubectl, create a Gemini API key, and run the kubectl-ai command in your terminal to start interacting with your cluster.

Key features

Key features of talk2-k8s include natural language processing for Kubernetes commands, integration with Gemini AI models, and the ability to validate user interactions through challenges.

Where to use

talk2-k8s can be used in cloud computing environments, DevOps practices, and any scenario where Kubernetes management and automation are required.

Content

Build with AI with Gemini: Let’s talk to your cluster - LLM Agents powered Kubernetes MCP Server

Resources

Pre Requisites

Setup

  1. Create Kubernetes cluster on Google Cloud
  2. Create a Gemini API key from (https://aistudio.google.com/apikey)
  3. Export the API key as an environment variable
export GEMINI_API_KEY=<API_KEY>

  1. Open a terminal and run the following command to start the kubectl-ai
kubectl-ai --model gemini-2.5-pro-exp-03-25  

or

kubectl-ai --model gemini-2.5-flash-preview-04-17

Additional Tips

Add a Kubernetes MCP server (Optional)

code --add-mcp '{"name":"kubernetes","command":"npx","args":["kubernetes-mcp-server@latest"]}'

Talk to Cluster Challenge

  1. Apply the following command to get started
kubectl apply -f https://raw.githubusercontent.com/chamodshehanka/talk2-k8s/refs/heads/main/bwai-manifests.yaml
  1. Complete the challenge by using the kubectl-ai
  2. Validate the challenge by running the following command
chmod +x ./validate.sh
./validate.sh

Submission Link: https://forms.gle/FdQaVMad1MszTzkM8

Tools

No tools

Comments

Recommend MCP Servers

View All MCP Servers