Supercharge Kubernetes with kubectl ai

Google’s kubectl-ai is a revolutionary open-source project that seamlessly integrates with Large Language Models (LLMs) like gemini, bedrock and local LLM providers such as ollama with Kubernetes operations, enabling users to interact with Kuberntes clusters using natural language.

🎞️ Watch Video

Pre-requisits

• Kubernetes cluster
• Kubectl is installed and configured
• Helm installaed and configured

Install ollama on your machine

curl -fsSL https://ollama.com/install.sh | sh

Verify

ollama --version

Pull a lightweight model — we’ll use Gemma 3 4‑B for the demo:

ollama pull gemma3:4b

kubectl‑ai

Ensure that you have kubectl is installed and configured. because kubectl-ai is a plugin that sits next to kubectl in your terminal.

curl -sSL https://raw.githubusercontent.com/GoogleCloudPlatform/kubectl-ai/main/install.sh | bash

Verify

kubectl ai version

Lets Ask our cluster a question :

kubectl ai --llm-provider ollama --model gemma3:4b --enable-tool-use-shim

You can also configure kubectl-ai using a YAML configuration file at ~/.config/kubectl-ai/config.yaml to persist those flags so you don’t have to re‑type them in.

{
mkdir -p ~/.config/kubectl-ai/
cat <<EOF > ~/.config/kubectl-ai/config.yaml
model: gemma3:4b
llmProvider: ollama
enableToolUseShim: true
toolConfigPaths: 
    - ~/.config/kubectl-ai/tools.yaml
EOF
}

verify your configuration

kubectl-ai model

We can Extend the Agent with Custom Tools

kubectl-ai can call any binary you describe in ~/.config/kubectl-ai/tools.yaml.

{
cat <<EOF > ~/.config/kubectl-ai/tools.yaml
- name: helm
  description: "Helm is the Kubernetes package manager."
  command: "helm"
  is_custom: true
EOF
}

To enable the custom tools, you must point kubectl-ai to the directory containing the tool configuration YAML files using the --custom-tools-config flag.

kubectl-ai --custom-tools-config=~/.config/kubectl-ai/tools.yaml "helm list"

Install Python and Create a Virtual Environment

First, make sure you have Python installed. You can verify this by running:

python3 --version

In the directory where you want your project, open a terminal and run:

python3 -m venv ollama-env

Activate the environment by running:

source ollama-env/bin/activate

Install the Ollama Python library:

pip3 install ollama

Run the python app

python3 python-app/kubectlai-app.py

🔗 Reference Links:

• kubectl-ai

• Github repo