Add StreamsHub MCP overlays

components/mcp/stack/kafka-access/kustomization.yaml

@@ -0,0 +1,13 @@
+apiVersion: kustomize.config.k8s.io/v1beta1
+kind: Kustomization
+
+namespace: kafka
+
+resources:
+  - https://raw.githubusercontent.com/streamshub/streamshub-mcp/v0.1.0/install/strimzi-mcp/optional/role-sensitive.yaml
+  - https://raw.githubusercontent.com/streamshub/streamshub-mcp/v0.1.0/install/strimzi-mcp/optional/rolebinding-sensitive.yaml
+
+labels:
+  - pairs:
+      app.kubernetes.io/part-of: streamshub-developer-quickstart
+    includeSelectors: false

components/mcp/stack/kustomization.yaml

@@ -0,0 +1,6 @@
+apiVersion: kustomize.config.k8s.io/v1alpha1
+kind: Component
+
+resources:
+  - strimzi-mcp
+  - kafka-access

components/mcp/stack/strimzi-mcp/kustomization.yaml

@@ -0,0 +1,14 @@
+apiVersion: kustomize.config.k8s.io/v1beta1
+kind: Kustomization
+
+resources:
+  - https://github.com/streamshub/streamshub-mcp//install/strimzi-mcp/base?ref=v0.1.0
+
+labels:
+  - pairs:
+      app.kubernetes.io/part-of: streamshub-developer-quickstart
+    includeSelectors: false
+
+images:
+  - name: quay.io/streamshub/strimzi-mcp
+    newTag: "0.1.0"

docs/overlays/mcp-metrics.md

@@ -0,0 +1,106 @@
++++
+title = 'MCP + Metrics'
+weight = 3
+cpu_total = '3 CPU cores'
+memory_total = '5.5 GiB'
++++
+
+The MCP + Metrics overlay combines the core stack with both Prometheus monitoring and the StreamsHub MCP server. 
+The MCP server is configured to use Prometheus as its centralized metrics provider.
+
+## Quick-Start Install
+
+```shell
+curl -sL https://raw.githubusercontent.com/streamshub/developer-quickstart/main/install.sh | OVERLAY=mcp-metrics bash
+```
+
+## Manual Install
+
+If you prefer step-by-step control, the MCP + Metrics overlay uses `overlays/mcp-metrics`:
+
+```shell
+# Phase 1 — Operators and CRDs (includes Prometheus Operator)
+kubectl apply --server-side --force-conflicts -k 'https://github.com/streamshub/developer-quickstart//overlays/mcp-metrics/base?ref=main'
+
+# Optionally, wait for the operators to be ready
+kubectl wait --for=condition=Available deployment/prometheus-operator -n monitoring --timeout=120s
+kubectl wait --for=condition=Available deployment/strimzi-cluster-operator -n strimzi --timeout=120s
+kubectl wait --for=condition=Available deployment/apicurio-registry-operator -n apicurio-registry --timeout=120s
+kubectl wait --for=condition=Available deployment/streamshub-console-operator -n streamshub-console --timeout=120s
+
+# Phase 2 — Operands, MCP server, Prometheus instance and monitors
+kubectl apply -k 'https://github.com/streamshub/developer-quickstart//overlays/mcp-metrics/stack?ref=main'
+```
+
+## Uninstall
+
+```shell
+curl -sL https://raw.githubusercontent.com/streamshub/developer-quickstart/main/uninstall.sh | OVERLAY=mcp-metrics bash
+```
+
+## What Gets Added
+
+On top of the components provided by the core overlay, the MCP + Metrics overlay adds:
+
+| Component              | Namespace        | Description                                              |
+|------------------------|------------------|----------------------------------------------------------|
+| Prometheus Operator    | `monitoring`     | Manages Prometheus instances and monitors                |
+| Prometheus instance    | `monitoring`     | Collects and stores metrics (400Mi memory, 1 replica)    |
+| PodMonitors            | `monitoring`     | Scrape targets for Kafka brokers and Strimzi operators   |
+| ServiceMonitor         | `monitoring`     | Scrape target for StreamsHub Console operator            |
+| StreamsHub MCP server  | `streamshub-mcp` | MCP server configured with Prometheus metrics provider   |
+| MCP PodMonitor         | `monitoring`     | Scrape target for MCP server metrics                     |
+| Sensitive Role         | `kafka`          | Grants access to TLS certificates and pod metrics        |
+
+The overlay patches existing resources:
+
+- Kafka — enables the [Strimzi Metrics Reporter](https://strimzi.io/docs/operators/latest/deploying#proc-metrics-kafka-str) on the `dev-cluster`
+- Console — adds Prometheus as a metrics data source so the Console UI displays Kafka metrics
+- MCP server — configured to use Prometheus as centralized metrics provider instead of direct pod scraping
+
+## Resource Requirements
+
+The MCP + Metrics overlay requires at least {{< param cpu_total >}} and {{< param memory_total >}} of allocatable cluster resources.
+This includes the resources for the [core](core.md) stack plus the additional components listed above.
+
+## Accessing the MCP Server
+
+Port-forward to the MCP server:
+
+```shell
+kubectl port-forward -n streamshub-mcp svc/streamshub-strimzi-mcp 8085:8080
+```
+
+The MCP endpoint is available at `http://localhost:8085/mcp`.
+
+## Accessing Prometheus
+
+Port-forward to the Prometheus UI:
+
+```shell
+kubectl port-forward -n monitoring svc/prometheus-operated 9090:9090
+```
+
+Open [http://localhost:9090](http://localhost:9090) and navigate to **Status > Targets** to verify all scrape targets are up, including the MCP server.
+
+## Verify the Installation
+
+Confirm all components are running:
+
+```shell
+kubectl get deployment -n streamshub-mcp streamshub-strimzi-mcp
+kubectl get prometheus -n monitoring
+kubectl get podmonitor -n monitoring
+```
+
+The MCP server deployment should show `1/1` ready replicas. Prometheus should reach `Available` status.
+
+## Troubleshooting
+
+### MCP Server Not Starting
+
+See the [MCP overlay troubleshooting](mcp.md#troubleshooting) section.
+
+### Metrics Not Appearing
+
+See the [Metrics overlay troubleshooting](metrics.md#metrics-not-appearing) section.

docs/overlays/mcp.md

@@ -0,0 +1,101 @@
++++
+title = 'MCP'
+weight = 2
+cpu_total = '3 CPU cores'
+memory_total = '5 GiB'
++++
+
+The MCP overlay extends the core stack with the StreamsHub MCP server, giving AI assistants (Claude Code, VS Code Copilot, etc.) read-only access to the Strimzi-managed Kafka cluster.
+
+## Quick-Start Install
+
+```shell
+curl -sL https://raw.githubusercontent.com/streamshub/developer-quickstart/main/install.sh | OVERLAY=mcp bash
+```
+
+## Manual Install
+
+If you prefer step-by-step control, the MCP overlay uses `overlays/mcp` instead of `overlays/core`:
+
+```shell
+# Phase 1 — Operators and CRDs
+kubectl apply --server-side --force-conflicts -k 'https://github.com/streamshub/developer-quickstart//overlays/mcp/base?ref=main'
+
+# Optionally, wait for the operators to be ready
+kubectl wait --for=condition=Available deployment/strimzi-cluster-operator -n strimzi --timeout=120s
+kubectl wait --for=condition=Available deployment/apicurio-registry-operator -n apicurio-registry --timeout=120s
+kubectl wait --for=condition=Available deployment/streamshub-console-operator -n streamshub-console --timeout=120s
+
+# Phase 2 — Operands and MCP server
+kubectl apply -k 'https://github.com/streamshub/developer-quickstart//overlays/mcp/stack?ref=main'
+```
+
+## Uninstall
+
+```shell
+curl -sL https://raw.githubusercontent.com/streamshub/developer-quickstart/main/uninstall.sh | OVERLAY=mcp bash
+```
+
+## What Gets Added
+
+On top of the components provided by the core overlay, the MCP overlay adds:
+
+| Component              | Namespace        | Description                                              |
+|------------------------|------------------|----------------------------------------------------------|
+| StreamsHub MCP server  | `streamshub-mcp` | MCP server for AI-assisted Kafka cluster management      |
+| Sensitive Role         | `kafka`          | Grants access to TLS certificates and pod metrics        |
+
+The MCP server provides read-only tools for inspecting and troubleshooting Strimzi-managed Kafka clusters via the [Model Context Protocol](https://modelcontextprotocol.io/).
+
+## Resource Requirements
+
+The MCP overlay requires at least {{< param cpu_total >}} and {{< param memory_total >}} of allocatable cluster resources.
+This includes the resources for the [core](core.md) stack plus the MCP server listed above.
+
+## Accessing the MCP Server
+
+Port-forward to the MCP server:
+
+```shell
+kubectl port-forward -n streamshub-mcp svc/streamshub-strimzi-mcp 8085:8080
+```
+
+The MCP endpoint is available at `http://localhost:8085/mcp`.
+
+## Verify the Installation
+
+Confirm the MCP server is running:
+
+```shell
+kubectl get deployment -n streamshub-mcp streamshub-strimzi-mcp
+```
+
+The deployment should show `1/1` ready replicas. Check the health endpoint:
+
+```shell
+kubectl port-forward -n streamshub-mcp svc/streamshub-strimzi-mcp 8085:8080 &
+curl -s http://localhost:8085/q/health/ready
+```
+
+## Troubleshooting
+
+### MCP Server Not Starting
+
+If the MCP server deployment is not becoming ready:
+
+```shell
+# Check pod status
+kubectl get pods -n streamshub-mcp
+
+# Check logs
+kubectl logs -n streamshub-mcp deployment/streamshub-strimzi-mcp
+
+# Verify Strimzi CRDs are installed
+kubectl get crd kafkas.kafka.strimzi.io
+```
+
+**Common causes:**
+
+- Strimzi operator not installed — the MCP server requires Strimzi CRDs to exist
+- Insufficient cluster resources — check node resource availability
+- Image pull errors — verify the cluster can pull from `quay.io`

docs/prerequisites.md

@@ -49,10 +49,10 @@ nodes:
             node-labels: "ingress-ready=true"
     extraPortMappings:
       - containerPort: 80
-        hostPort: 80
+        hostPort: 8080
         protocol: TCP
       - containerPort: 443
-        hostPort: 443
+        hostPort: 8443
         protocol: TCP
 EOF
 ```

install.sh

@@ -88,10 +88,19 @@ main() {
         stack_path="overlays/${OVERLAY}/stack"
     fi
 
+    # Determine which optional features are included in this overlay
+    local has_metrics=false
+    local has_mcp=false
+    case "${OVERLAY}" in
+        metrics)     has_metrics=true ;;
+        mcp)         has_mcp=true ;;
+        mcp-metrics) has_metrics=true; has_mcp=true ;;
+    esac
+
     # Compute total steps based on overlay and verification
     local total_steps=5
-    if [ "${OVERLAY}" = "metrics" ]; then
-        total_steps=6
+    if [ "$has_metrics" = true ]; then
+        total_steps=$((total_steps + 1))
     fi
     if [ -z "${SKIP_VERIFY}" ]; then
         total_steps=$((total_steps + 1))
@@ -127,8 +136,8 @@ main() {
     kubectl apply --server-side --force-conflicts -k "${base_url}"
     echo ""
 
-    # --- Step: Wait for prometheus-operator (metrics overlay only) ---
-    if [ "${OVERLAY}" = "metrics" ]; then
+    # --- Step: Wait for prometheus-operator (metrics overlays only) ---
+    if [ "$has_metrics" = true ]; then
         step=$((step + 1))
         info "Step ${step}/${total_steps}: Waiting for prometheus-operator to be ready (timeout: ${TIMEOUT})..."
         kubectl wait --for=condition=Available deployment/prometheus-operator \
@@ -203,7 +212,7 @@ main() {
             verify_failed=true
         fi
 
-        if [ "${OVERLAY}" = "metrics" ]; then
+        if [ "$has_metrics" = true ]; then
             info "  Waiting for Prometheus..."
             if kubectl wait --for=condition=Available prometheus.monitoring.coreos.com/prometheus \
                     -n monitoring --timeout="${TIMEOUT}" 2>/dev/null; then
@@ -214,6 +223,17 @@ main() {
             fi
         fi
 
+        if [ "$has_mcp" = true ]; then
+            info "  Waiting for StreamsHub MCP server..."
+            if kubectl wait --for=condition=Available deployment/streamshub-strimzi-mcp \
+                    -n streamshub-mcp --timeout="${TIMEOUT}" 2>/dev/null; then
+                info "  StreamsHub MCP server is ready"
+            else
+                warn "  StreamsHub MCP server did not become ready within ${TIMEOUT}"
+                verify_failed=true
+            fi
+        fi
+
         echo ""
     fi
 
@@ -234,26 +254,38 @@ main() {
     echo "  - Apicurio Registry instance  (namespace: apicurio-registry, storage: in-memory)"
     echo "  - StreamsHub Console operator (namespace: streamshub-console)"
     echo "  - StreamsHub Console instance (namespace: streamshub-console)"
-    if [ "${OVERLAY}" = "metrics" ]; then
+    if [ "$has_metrics" = true ]; then
         echo "  - Prometheus operator         (namespace: monitoring)"
         echo "  - Prometheus instance          (namespace: monitoring)"
         echo "  - Kafka metrics (PodMonitors) (namespace: monitoring)"
     fi
+    if [ "$has_mcp" = true ]; then
+        echo "  - StreamsHub MCP server       (namespace: streamshub-mcp)"
+    fi
     echo ""
 
     if [ -z "${SKIP_VERIFY}" ] && [ "${verify_failed}" != "true" ]; then
         echo "Access the Console:"
         echo "  kubectl port-forward -n streamshub-console svc/streamshub-console-console-service 8090:80"
         echo "  Then open http://localhost:8090"
+        if [ "$has_mcp" = true ]; then
+            echo ""
+            echo "Access the MCP server:"
+            echo "  kubectl port-forward -n streamshub-mcp svc/streamshub-strimzi-mcp 8085:8080"
+            echo "  MCP endpoint: http://localhost:8085/mcp"
+        fi
         echo ""
     elif [ -z "${SKIP_VERIFY}" ] && [ "${verify_failed}" = "true" ]; then
         echo "Check operand status:"
         echo "  kubectl get kafka -n kafka"
         echo "  kubectl get apicurioregistry3 -n apicurio-registry"
         echo "  kubectl get console -n streamshub-console"
-        if [ "${OVERLAY}" = "metrics" ]; then
+        if [ "$has_metrics" = true ]; then
             echo "  kubectl get prometheus -n monitoring"
         fi
+        if [ "$has_mcp" = true ]; then
+            echo "  kubectl get deployment -n streamshub-mcp streamshub-strimzi-mcp"
+        fi
         echo ""
         echo "Some resources may still be starting up. Re-check after a few minutes."
         echo ""
@@ -265,10 +297,13 @@ main() {
         echo "  kubectl get apicurioregistry3 -n apicurio-registry"
         echo "  kubectl get deployment -n streamshub-console console-operator"
         echo "  kubectl get console -n streamshub-console"
-        if [ "${OVERLAY}" = "metrics" ]; then
+        if [ "$has_metrics" = true ]; then
             echo "  kubectl get prometheus -n monitoring"
             echo "  kubectl get podmonitor -n monitoring"
         fi
+        if [ "$has_mcp" = true ]; then
+            echo "  kubectl get deployment -n streamshub-mcp streamshub-strimzi-mcp"
+        fi
         echo ""
         echo "Note: It may take several minutes for all resources to become ready."
         echo ""

overlays/mcp-metrics/base/kustomization.yaml

@@ -0,0 +1,5 @@
+apiVersion: kustomize.config.k8s.io/v1beta1
+kind: Kustomization
+
+resources:
+  - ../../metrics/base

overlays/mcp-metrics/stack/kustomization.yaml

@@ -0,0 +1,18 @@
+apiVersion: kustomize.config.k8s.io/v1beta1
+kind: Kustomization
+
+components:
+  - ../../../components/core/stack
+  - ../../../components/metrics/stack
+  - ../../../components/mcp/stack
+
+resources:
+  - mcp-podmonitor.yaml
+
+patches:
+  - target:
+      group: apps
+      version: v1
+      kind: Deployment
+      name: streamshub-strimzi-mcp
+    path: mcp-prometheus-patch.yaml