Agent Runtime Operator Documentation
The Agent Runtime Operator provides a unified, framework-agnostic way to deploy and manage agentic workloads on Kubernetes. It uses Custom Resource Definitions (CRDs) to abstract away framework-specific configurations, simplifying the management of AI agent deployments.
How-To Guides
Step-by-step guides to get you up and running:
-
Install the Agent Runtime Operator - Complete installation guide with prerequisites, setup steps, and verification
-
Create an AI Agent - Learn how to deploy template-based and custom agents with step-by-step examples
Advanced Topics
For advanced routing and gateway patterns:
-
Create a KrakenD Agent Gateway - Set up agent gateways to manage routing to agents
-
Create a LiteLLM AI Gateway - Set up AI gateways to manage routing from agents to LLMs
Reference
Detailed reference information for all components:
-
Agent Runtime Reference - Complete API reference for all custom resources
External Resources
-
Agent Samples - Example implementations and starter templates
-
Python SDK - SDK for building ADK-based agents
-
Source Code - Main repository and issue tracking