Deploy Kubernetes-native agents with kagent
Define agents, model configs, and MCP tool servers as Kubernetes resources so cloud operators can run controlled infrastructure workflows in-cluster.
npx skills add agentskillexchange/skills --skill deploy-kubernetes-native-agents-with-kagent
Use kagent when the agent workflow belongs inside Kubernetes operations. The operator installs kagent, defines an Agent resource with a system prompt, model configuration, and tool servers, then runs tasks against Kubernetes, Istio, Helm, Argo, Prometheus, Grafana, Cilium, or other connected infrastructure tools while inspecting execution and traces. The skill boundary is Kubernetes-native agent operation through CRDs and MCP-backed tool servers; avoid using it as a generic agent framework listing.
What this skill actually does
Inputs and prerequisites: Kubernetes cluster, kubectl, Helm, kagent, MCP tool servers, supported LLM provider.
Setup notes: Follow the kagent installation guide, install kagent into a Kubernetes cluster, configure a ModelConfig for the selected provider, define ToolServers for the infrastructure tools the agent may use, and create an Agent resource for the bounded operations workflow.
Source and verification boundary: use https://kagent.dev/docs/kagent/getting-started/quickstart as the canonical reference before running the workflow; keep commands, API calls, CLI usage, and generated outputs reviewable against that upstream source.
Framework fit: publish this as a Multi-Framework workflow only when the operator can invoke the documented toolchain directly, rather than treating the upstream project as a generic product listing.