InstallKubeiQ

Cluster Doctor with a root cause and the commands to fix itCluster Doctor with a root cause and the commands to fix it

One command installs every KubeiQ product on your laptop or server

  • curl -fsSL https://raw.githubusercontent.com/Ahmedwagdymohy/KubeiQ/main/deploy/install.sh | sh
The KubeiQ Overview you open after the install

Five steps to this Overview, the same on a laptop or a server

  1. Check the Requirements

    • Docker with Compose v2, running
    • git
    • A kubeconfig at ~/.kube/config with access to your clusters. KubeiQ only needs read access

  2. Run the Installer

    • curl -fsSL https://raw.githubusercontent.com/Ahmedwagdymohy/KubeiQ/main/deploy/install.sh | sh

    It downloads KubeiQ to ~/kubeiq, builds it and starts it with Docker Compose. The first run takes a few minutes.

    Port 3000 already in use? Choose another one:

    • export KUBEIQ_FRONTEND_PORT=3001
    • curl -fsSL https://raw.githubusercontent.com/Ahmedwagdymohy/KubeiQ/main/deploy/install.sh | sh

  3. Open KubeiQ

    Go to http://localhost:3000 and sign in as admin with the master key the installer prints. Open Cluster Doctor, pick a cluster and a namespace, and start a scan.

    Cluster Doctor after a scan, showing a failing workload and the fixCluster Doctor after a scan, showing a failing workload and the fix

  4. Add an AI Model

    Optional. Without a model, KubeiQ still finds known failures and gives you the fix. With one, it also investigates the unclear cases. Add it in Settings, or set it in ~/kubeiq/backend/.env, for example:

    • KUBEIQ_LLM_MODEL=openai/gpt-4.1
    • KUBEIQ_LLM_API_KEY=<your-api-key>

    Then apply it:

    • cd ~/kubeiq && docker compose up -d

    Anthropic, OpenRouter, Azure, Bedrock, Gemini, Ollama and LiteLLM proxies work too. The file lists an example for each.

  5. Update or Remove

    To update, run the installer again. To stop KubeiQ:

    • cd ~/kubeiq && docker compose down

Install Manually

Prefer to see every step? The installer runs exactly this:

  • git clone https://github.com/Ahmedwagdymohy/KubeiQ.git ~/kubeiq
  • cd ~/kubeiq && cp backend/.env.example backend/.env
  • docker compose up --build -d

Kubernetes

A Helm chart and an in-cluster agent for private clusters are on the way. Follow the docs for updates.