Omnia

Deploy and manage AI agents on Kubernetes with ease

What is Omnia?

Omnia is a Kubernetes operator that simplifies the deployment and management of AI agents. It provides custom resources for defining agent configurations, prompt packs, and tool registries, enabling you to deploy intelligent agents at scale.

AgentRuntime CRD

Deploy AI agents as Kubernetes-native workloads with automatic scaling and health management.

PromptPack CRD

Version and manage prompt configurations with canary rollouts for safe updates.

ToolRegistry CRD

Discover and orchestrate tools via HTTP, gRPC, MCP, or OpenAPI handlers.

WebSocket Facade

Real-time communication with deployed agents through a streaming interface.

Quick Start

Install Omnia using Helm:

helm repo add omnia https://altairalabs.github.io/omnia/charts
helm install omnia omnia/omnia --namespace omnia-system --create-namespace

Then deploy your first agent:

apiVersion: omnia.altairalabs.ai/v1alpha1
kind: AgentRuntime
metadata:
  name: my-agent
spec:
  replicas: 1
  providerRef:
    name: my-provider
  promptPackRef:
    name: my-promptpack

For a complete walkthrough, see the Getting Started Tutorial.

Documentation Structure

This documentation follows the Diataxis framework:

Tutorials

Step-by-step guides to get you started with Omnia. Start Learning →

How-To Guides

Practical guides for specific tasks and goals. Find Solutions →

Reference

Technical specifications, CRD schemas, and API docs. Browse Reference →

Concepts

Background and explanations of how Omnia works. Understand Omnia →