# KodeKloud: Beta Courses KAgent: Host Your AI Agents on Kubernetes

## KAgent: Host Your AI Agents on Kubernetes

### Introduction

- [Course Introduction](https://notes.kodekloud.com/docs/KAgent-Host-Your-AI-Agents-on-Kubernetes/Introduction/Course-Introduction/page.md): Hands-on course teaching deployment, management, and observability of AI agents on Kubernetes with KAgent, KMCP/MCP, system prompts, and integrations like Slack and AWS
- [Why Use KAgent](https://notes.kodekloud.com/docs/KAgent-Host-Your-AI-Agents-on-Kubernetes/Introduction/Why-Use-KAgent/page.md): KAgent is an open source framework for running agentic AI on Kubernetes, offering CRDs, controllers, multi‑LLM support, observability, and GitOps-friendly deployment for DevOps and platform teams.
- [Cons and Limitations of KAgent](https://notes.kodekloud.com/docs/KAgent-Host-Your-AI-Agents-on-Kubernetes/Introduction/Cons-and-Limitations-of-KAgent/page.md): Overview of KAgent limitations, operational requirements, and technical trade-offs for hosting AI agents on Kubernetes, with guidance on suitability and deployment considerations
- [Demo KAgent Overview Of UI and Agents](https://notes.kodekloud.com/docs/KAgent-Host-Your-AI-Agents-on-Kubernetes/Introduction/Demo-KAgent-Overview-Of-UI-and-Agents/page.md): Overview of KAgent web UI showing agents, models, tools, and MCP server integrations for managing AI-driven Kubernetes automation and configuring LLM providers and tool actions
- [Demo Standard K8s Issues and Explore MCP Tools](https://notes.kodekloud.com/docs/KAgent-Host-Your-AI-Agents-on-Kubernetes/Introduction/Demo-Standard-K8s-Issues-and-Explore-MCP-Tools/page.md): Hands-on lab demonstrating troubleshooting of Kubernetes service selector and HPA quota issues, comparing manual kubectl diagnostics with AI-assisted KAgent remediation.

### KAgent Installation & Architecture Overview

- [Agent to Agent A2A Protocol](https://notes.kodekloud.com/docs/KAgent-Host-Your-AI-Agents-on-Kubernetes/KAgent-Installation-Architecture-Overview/Agent-to-Agent-A2A-Protocol/page.md): Explains A2A, an open standard enabling secure, discoverable, interoperable agent-to-agent communication for discovery, invocation, security, streaming, and long-running multi-agent collaboration.
- [Key Design Princiles of A2A](https://notes.kodekloud.com/docs/KAgent-Host-Your-AI-Agents-on-Kubernetes/KAgent-Installation-Architecture-Overview/Key-Design-Princiles-of-A2A/page.md): Overview of Google's A2A protocol design principles enabling secure, interoperable, enterprise-ready agent to agent communication with flexible patterns, modality independence, opaque execution, and framework agnosticism.
- [Fundamental Communication Elements](https://notes.kodekloud.com/docs/KAgent-Host-Your-AI-Agents-on-Kubernetes/KAgent-Installation-Architecture-Overview/Fundamental-Communication-Elements/page.md): Defines core elements and flow for agent-to-agent communication including agent cards, messages and parts, tasks with streaming, artifacts as deliverables, discovery, transport, and authentication
- [Agent Discovery](https://notes.kodekloud.com/docs/KAgent-Host-Your-AI-Agents-on-Kubernetes/KAgent-Installation-Architecture-Overview/Agent-Discovery/page.md): Describes agent card discovery methods for A2A systems, comparing well-known URIs, registries, and direct configuration, plus security practices and deployment trade-offs.
- [A2A Request Lifecycle](https://notes.kodekloud.com/docs/KAgent-Host-Your-AI-Agents-on-Kubernetes/KAgent-Installation-Architecture-Overview/A2A-Request-Lifecycle/page.md): Explains the agent-to-agent request lifecycle, covering discovery, authentication, sendMessage and sendMessageStream for secure interoperable multi-agent communication and streaming task updates.
- [KAgent Architecture](https://notes.kodekloud.com/docs/KAgent-Host-Your-AI-Agents-on-Kubernetes/KAgent-Installation-Architecture-Overview/KAgent-Architecture/page.md): Describes KAgent architecture that integrates LLM-driven autonomous agents, cloud-native tools, and a Kubernetes-native runtime to automate DevOps tasks, observability, and troubleshooting.
- [Agents](https://notes.kodekloud.com/docs/KAgent-Host-Your-AI-Agents-on-Kubernetes/KAgent-Installation-Architecture-Overview/Agents/page.md): Explains AI agent architecture, detailing instructions, tools, skills, and best practices for designing and managing agents in KAgent
- [Tools](https://notes.kodekloud.com/docs/KAgent-Host-Your-AI-Agents-on-Kubernetes/KAgent-Installation-Architecture-Overview/Tools/page.md): Overview of K-Agent tools, MCP integration, and agent-as-tool composition enabling Kubernetes-native agents to call built-in and external tools for cluster, monitoring, and cloud operations.
- [Demo KAgent Installation](https://notes.kodekloud.com/docs/KAgent-Host-Your-AI-Agents-on-Kubernetes/KAgent-Installation-Architecture-Overview/Demo-KAgent-Installation/page.md): Guide to installing and uninstalling kagent on Kubernetes using either the kagent CLI installer or Helm charts, including prerequisites and verification steps
- [Demo KAgent Exploration](https://notes.kodekloud.com/docs/KAgent-Host-Your-AI-Agents-on-Kubernetes/KAgent-Installation-Architecture-Overview/Demo-KAgent-Exploration/page.md): Step-by-step guide to install and configure KAgent on Kubernetes, manage ModelConfig and provider secrets, enable tools and k8s-agent, and interact via UI and kagent CLI.

### KMCP Installation & Overview 

- [Kubernetes Model Context KMCP](https://notes.kodekloud.com/docs/KAgent-Host-Your-AI-Agents-on-Kubernetes/KMCP-Installation-Overview/Kubernetes-Model-Context-KMCP/page.md): KMCP is a toolkit and CLI for scaffolding, testing, and deploying MCP servers in Kubernetes, providing boilerplates, local inspection, CRD lifecycle and secret management.
- [Demo KMCP Installation](https://notes.kodekloud.com/docs/KAgent-Host-Your-AI-Agents-on-Kubernetes/KMCP-Installation-Overview/Demo-KMCP-Installation/page.md): Guide to installing and verifying the KMCP controller bundled with KAgent, checking CRDs and resources, inspecting components, and deploying a sample MCPServer to validate reconciliation.
- [Demo Installing AWS Servers on KBs via KAgent UI](https://notes.kodekloud.com/docs/KAgent-Host-Your-AI-Agents-on-Kubernetes/KMCP-Installation-Overview/Demo-Installing-AWS-Servers-on-KBs-via-KAgent-UI/page.md): Guide to installing, configuring, and verifying two AWS MCP servers on KAgent using the UI and a YAML manifest, including credential setup and troubleshooting.
- [Demo Create AI Agents Using AWS MCP Servers In action](https://notes.kodekloud.com/docs/KAgent-Host-Your-AI-Agents-on-Kubernetes/KMCP-Installation-Overview/Demo-Create-AI-Agents-Using-AWS-MCP-Servers-In-action/page.md): Lab walkthrough to deploy declarative AI agents with AWS MCP servers and KAgent UI to query real-time AWS pricing, attach pricing tools, run queries, and troubleshoot
- [Demo Deploying MCP Server KMCP Way](https://notes.kodekloud.com/docs/KAgent-Host-Your-AI-Agents-on-Kubernetes/KMCP-Installation-Overview/Demo-Deploying-MCP-Server-KMCP-Way/page.md): Guide to scaffolding, running, debugging, building, and deploying a Python MCP server with KMCP and MCP Inspector exposing cryptocurrency price tools
- [Demo Deploying MCP Server KMCP Way Part 2](https://notes.kodekloud.com/docs/KAgent-Host-Your-AI-Agents-on-Kubernetes/KMCP-Installation-Overview/Demo-Deploying-MCP-Server-KMCP-Way-Part-2/page.md): Guide to adding a crypto price MCP tool, building and deploying an MCP server and declarative agent, and testing tool usage via MCP Inspector and KAgent UI

### System Prompts for AI Agent Building

- [AI Agent System Prompts Core Concepts](https://notes.kodekloud.com/docs/KAgent-Host-Your-AI-Agents-on-Kubernetes/System-Prompts-for-AI-Agent-Building/AI-Agent-System-Prompts-Core-Concepts/page.md): Explains core concepts and best practices for designing system prompts that define agent instructions, tool usage, and user query handling to build reliable, safe AI agents
- [Prompt Structure](https://notes.kodekloud.com/docs/KAgent-Host-Your-AI-Agents-on-Kubernetes/System-Prompts-for-AI-Agent-Building/Prompt-Structure/page.md): Guide to designing clear actionable system prompts for Kubernetes management agents, covering tools, parameters, operational protocols, safety guardrails, and confirmation rules
- [Prompt Structure Components](https://notes.kodekloud.com/docs/KAgent-Host-Your-AI-Agents-on-Kubernetes/System-Prompts-for-AI-Agent-Building/Prompt-Structure-Components/page.md): Guide for designing AI agent prompts and configurations covering role, goals, tools, operational protocols, safety, testing, and storage best practices.

### Debugging AI Agents

- [Demo Debugging AI Agents](https://notes.kodekloud.com/docs/KAgent-Host-Your-AI-Agents-on-Kubernetes/Debugging-AI-Agents/Demo-Debugging-AI-Agents/page.md): Guide showing how to configure Jaeger and KAgent to export OpenTelemetry traces, generate agent traffic, and inspect AI agent traces and LLM metadata in Jaeger

### Creating AI Agents - Bring It all together

- [Demo Create AI Agents using AWS MCP Servers In action](https://notes.kodekloud.com/docs/KAgent-Host-Your-AI-Agents-on-Kubernetes/Creating-AI-Agents-Bring-It-all-together/Demo-Create-AI-Agents-using-AWS-MCP-Servers-In-action/page.md): Guide demonstrating two-way Slack and KAgent integration using A2A and Slack MCP servers, showing local Slack bot setup, agent deployment, tool configuration, and example workflows