Documentation

Introducing kagent

Kagent is an open-source programming framework that brings the power of agentic AI to cloud-native environments. Built specifically for DevOps and platform engineers, Kagent enables AI agents to run directly in Kubernetes clusters to automate operations, troubleshoot issues, and solve complex cloud-native challenges.

What is kagent?

Unlike traditional chatbots, kagent leverages advanced reasoning and iterative planning capabilities to autonomously handle multi-step problems in cloud-native environments. It transforms AI insights into concrete actions, helping teams tackle common operational challenges like:

  • Diagnosing connectivity issues across multiple service hops
  • Troubleshooting application performance degradation
  • Automating alert generation from Prometheus metrics
  • Debugging Gateway and HTTPRoute configurations
  • Managing progressive rollouts with Argo Rollouts

Core Components

Kagent's architecture consists of three main components:

  • Tools: Any MCP-style function that agents can leverage to interact with cloud-native systems. Kagent comes with pre-built tools that include capabilities like displaying pod logs, querying Prometheus metrics, generating resources and more. You can check the available tools in the tool registry.
  • Agents: Autonomous systems that plan, execute, and analyze tasks using the available tools. These agents can chain multiple operations together to solve complex problems. Each agent can have access to one or more tools to accomplish its work. Agents can also be grouped into teams where a planning agent comes up with a plan and assigns tasks to individual agents in the team.
  • Framework: A flexible interface that allows running agents either through a UI or declaratively. Built on Microsoft's AutoGen framework, it provides extensive customization options.

Why kagent?

Kagent addresses the growing complexity of cloud-native operations by:

  • Automating routine troubleshooting and operational tasks
  • Reducing the need for specialist intervention in common scenarios
  • Enabling teams to codify and share their operational expertise
  • Providing a platform for building and sharing custom AI agents

Getting Started

To start using kagent in your environment, check out the Quick Start Guide guide. For a deeper understanding of how kagent works, refer to the kagent architecture.

Ready to contribute? Visit our Github repository to learn how you can help expand the ecosystem of cloud-native AI agents.

Community

Join the kagent community:

  • Explore our repositories on GitHub
  • Join the discussion in the #kagent channel on CNCF Slack
  • Check our FAQ for common questions
  • Follow our Feature Roadmap for upcoming developments