kavryctl
CLI for registering, inspecting, and testing MCP server definitions.
Kavrynt helps platform teams register, operate, and route production MCP servers for AI agents on Kubernetes. Kavrynt is a private commercial SaaS product with approved alpha and beta trial images for developer evaluation.
Developers can define an MCP server, sync it into Registry, and route traffic through Gateway on any Kubernetes cluster. Kind is only the local test option.
CLI for registering, inspecting, and testing MCP server definitions.
Source of truth for MCP server metadata and routeable endpoints.
Data-plane entry point that routes AI client traffic to MCP servers.
Kubernetes controller that syncs MCPServer resources into Registry.
Developers evaluate Kavrynt by pulling approved trial images into their own cluster. Kind is useful when you want a disposable local cluster on your laptop.
export KAVRYNT_IMAGE_REGISTRY=docker.io/kavrynt
export KAVRYNT_TRIAL_TAG=0.1.0-beta
kubectl config current-context
kubectl get nodes
# Local testing only
# kind create cluster --name kavrynt-dev
docker pull "$KAVRYNT_IMAGE_REGISTRY/registry:$KAVRYNT_TRIAL_TAG"
docker pull "$KAVRYNT_IMAGE_REGISTRY/gateway:$KAVRYNT_TRIAL_TAG"
docker pull "$KAVRYNT_IMAGE_REGISTRY/k8s-operator:$KAVRYNT_TRIAL_TAG"
kubectl create namespace kavrynt-system
kubectl get pods -n kavrynt-systemInstalls kavryctl and declares MCP server intent.
CLIRegistry and Operator keep server metadata in sync.
KavryntGateway exposes stable MCP routing for AI clients.
MCPKavrynt is ready for early MVP testing by developers who want to run MCP infrastructure on Kubernetes, from local Kind clusters to managed cloud clusters.