Does this skill require an existing Kubernetes cluster?

No, the deploy skill automatically creates a k3d cluster (Kubernetes running in Podman) if one does not already exist when you run the config command.

What applications can I deploy with this skill?

It currently supports automated deployment for JupyterHub (multi-user notebook servers) and KubeAI (GPU-accelerated LLM inference).

How does it handle multiple environments?

The skill uses an --instance parameter to manage separate, isolated clusters and configurations, allowing you to run multiple deployments on the same machine.

Do I need to install Helm manually?

Helm is a prerequisite for this skill. While it's not included in the base plugin, you can easily install it using the provided 'ujust install helm' command.

Kubernetes App Deployment Manager

Name: Kubernetes App Deployment Manager
Author: atrawog

byatrawog

배포 및 DevOps

Manages Helm-based application deployments to local k3d Kubernetes clusters for AI and notebook workloads.

소개

This skill automates the full lifecycle of deploying specialized applications like JupyterHub and KubeAI onto local k3d Kubernetes clusters within the Bazzite AI ecosystem. It handles infrastructure orchestration—including automatic cluster creation using Podman—configuration management via Helm values, and provides a unified interface for scaling, monitoring, and upgrading multi-user environments and GPU-accelerated inference servers. It is ideal for developers and data scientists who need to manage local cloud-native AI infrastructure without manual Kubernetes boilerplate.

주요 기능

Streamlined Helm-based deployment for JupyterHub and KubeAI
Built-in status monitoring and ingress management for local access
0 GitHub stars
Multi-instance support for running isolated cluster environments
Configurable resource limits for CPU, memory, and GPU acceleration
Automated k3d Kubernetes cluster creation and lifecycle management

사용 사례

Deploying GPU-accelerated LLM inference servers with KubeAI and OpenAI-compatible APIs
Setting up a local multi-user JupyterHub environment for data science teams
Testing and debugging Kubernetes-native applications in a lightweight local k3s environment

소개

주요 기능

Streamlined Helm-based deployment for JupyterHub and KubeAI
Built-in status monitoring and ingress management for local access
0 GitHub stars
Multi-instance support for running isolated cluster environments
Configurable resource limits for CPU, memory, and GPU acceleration
Automated k3d Kubernetes cluster creation and lifecycle management

사용 사례

Deploying GPU-accelerated LLM inference servers with KubeAI and OpenAI-compatible APIs
Setting up a local multi-user JupyterHub environment for data science teams
Testing and debugging Kubernetes-native applications in a lightweight local k3s environment