diff --git a/README.md b/README.md
index 8b53eb1..c84d17a 100644
--- a/README.md
+++ b/README.md
@@ -399,6 +399,26 @@ _Serve mode_
 k8sgpt serve
 ```
 
+_Serve mode with MCP (Model Context Protocol)_
+
+```
+# Enable MCP server on default port 8089
+k8sgpt serve --mcp --mcp-http
+
+# Enable MCP server on custom port
+k8sgpt serve --mcp --mcp-http --mcp-port 8089
+
+# Full serve mode with MCP
+k8sgpt serve --mcp --mcp-http --port 8080 --metrics-port 8081 --mcp-port 8089
+```
+
+The MCP server enables integration with tools like Claude Desktop and other MCP-compatible clients. It runs on port 8089 by default and provides:
+- Kubernetes cluster analysis via MCP protocol
+- Resource information and health status
+- AI-powered issue explanations and recommendations
+
+For Helm chart deployment with MCP support, see the `charts/k8sgpt/values-mcp-example.yaml` file.
+
 _Analysis with serve mode_
 
 ```
diff --git a/charts/k8sgpt/Chart.yaml b/charts/k8sgpt/Chart.yaml
index a6f56e0..38fe55b 100644
--- a/charts/k8sgpt/Chart.yaml
+++ b/charts/k8sgpt/Chart.yaml
@@ -1,5 +1,5 @@
 apiVersion: v2
-appVersion: v0.3.0 #x-release-please-version
+appVersion: v0.4.23 #x-release-please-version
 description: A Helm chart for K8SGPT
 name: k8sgpt
 type: application
diff --git a/charts/k8sgpt/templates/deployment.yaml b/charts/k8sgpt/templates/deployment.yaml
index fcba48e..13c1e01 100644
--- a/charts/k8sgpt/templates/deployment.yaml
+++ b/charts/k8sgpt/templates/deployment.yaml
@@ -32,7 +32,13 @@ spec:
         image: {{ .Values.deployment.image.repository }}:{{ .Values.deployment.image.tag | default .Chart.AppVersion }}
         ports:
         - containerPort: 8080
-        args: ["serve"]
+        {{- if .Values.deployment.mcp.enabled }}
+        - containerPort: {{ .Values.deployment.mcp.port | int }}
+        {{- end }}
+        args: ["serve"
+        {{- if .Values.deployment.mcp.enabled }}, "--mcp", "-v","--mcp-http", "--mcp-port", {{ .Values.deployment.mcp.port | quote }}
+        {{- end }}
+        ]
         {{- if .Values.deployment.resources }}
         resources:
         {{- toYaml .Values.deployment.resources | nindent 10 }}
diff --git a/charts/k8sgpt/templates/service.yaml b/charts/k8sgpt/templates/service.yaml
index 6491289..c1d83a4 100644
--- a/charts/k8sgpt/templates/service.yaml
+++ b/charts/k8sgpt/templates/service.yaml
@@ -19,4 +19,9 @@ spec:
     - name: metrics
       port: 8081
       targetPort: 8081
+    {{- if .Values.deployment.mcp.enabled }}
+    - name: mcp
+      port: {{ .Values.deployment.mcp.port | int }}
+      targetPort: {{ .Values.deployment.mcp.port | int }}
+    {{- end }}
   type: {{ .Values.service.type }}
diff --git a/charts/k8sgpt/values-mcp-example.yaml b/charts/k8sgpt/values-mcp-example.yaml
new file mode 100644
index 0000000..5c97bc3
--- /dev/null
+++ b/charts/k8sgpt/values-mcp-example.yaml
@@ -0,0 +1,39 @@
+# Example values file to enable MCP (Model Context Protocol) service
+# Copy this file and modify as needed, then use: helm install -f values-mcp-example.yaml
+
+deployment:
+  # Enable MCP server
+  mcp:
+    enabled: true
+    port: "8089"  # Port for MCP server (default: 8089)
+    http: true    # Enable HTTP mode for MCP server
+  
+  # Other deployment settings remain the same
+  image:
+    repository: ghcr.io/k8sgpt-ai/k8sgpt
+    tag: "" # defaults to Chart.appVersion if unspecified
+  imagePullPolicy: Always
+  env:
+    model: "gpt-3.5-turbo"
+    backend: "openai"
+  resources:
+    limits:
+      cpu: "1"
+      memory: "512Mi"
+    requests:
+      cpu: "0.2"
+      memory: "156Mi"
+
+# Service configuration
+service:
+  type: ClusterIP
+  annotations: {}
+
+# Secret configuration for AI backend
+secret:
+  secretKey: "" # base64 encoded OpenAI token
+
+# ServiceMonitor for Prometheus metrics
+serviceMonitor:
+  enabled: false
+  additionalLabels: {}
diff --git a/charts/k8sgpt/values.yaml b/charts/k8sgpt/values.yaml
index 7a48bd1..9f5b74f 100644
--- a/charts/k8sgpt/values.yaml
+++ b/charts/k8sgpt/values.yaml
@@ -7,6 +7,11 @@ deployment:
   env:
     model: "gpt-3.5-turbo"
     backend: "openai" # one of: [ openai | llama ]
+  # MCP (Model Context Protocol) server configuration
+  mcp:
+    enabled: false # Enable MCP server
+    port: "8089" # Port for MCP server
+    http: true # Enable HTTP mode for MCP server
   resources:
     limits:
       cpu: "1"
diff --git a/pkg/server/mcp.go b/pkg/server/mcp.go
index e9a7b36..28aa463 100644
--- a/pkg/server/mcp.go
+++ b/pkg/server/mcp.go
@@ -141,6 +141,8 @@ func (s *K8sGptMCPServer) registerToolsAndResources() error {
 		),
 		mcp.WithArray("filters",
 			mcp.Description("Provide filters to narrow down the analysis (e.g. ['Pods', 'Deployments'])"),
+			// without below line MCP server fails with Google Agent Development Kit (ADK), interestingly works fine with mcpinspector
+			mcp.WithStringItems(),
 		),
 	)
 	s.server.AddTool(analyzeTool, s.handleAnalyze)