Merge pull request #166 from papagala/main

Updating docs to make the up to date and fixed bugs

Author: peterj

public/images/add-tool-server.png

@@ -79,21 +79,41 @@ docker build -t <your docker repository> .
 docker push <your docker repository> .
 ```
 
+> ** Important: Dockerfile Modifications Required**
+> 
+> Make sure you add these essential lines to the Dockerfile before building:
+> 
+> ```dockerfile
+> # Add this environment variable
+> ENV SQLITE_DB_DIR=/data
+> 
+> # Create the data directory and copy your database
+> RUN mkdir -p /data
+> COPY mcp.db /data/mcp.db
+> 
+> # Update the ownership to include /data
+> RUN chown -R kagent:nodejs /app /data
+> ```
+> 
+> These changes ensure the MCP server can locate your custom documentation database in the `/data` directory.
+
 ## Deploy the MCP server
 
 Create a Secret for the OpenAI API Key which is going to be used to create the embeddings:
 
 ```bash
 kubectl create secret generic mcp-secrets \
-  --from-literal=OPENAI_API_KEY=<your_openai_api_key>
+  --from-literal=OPENAI_API_KEY=<your_openai_api_key> \
+  -n kagent
 ```
 
 Create a ConfigMap for the Database Configuration:
 
 ```bash
 kubectl create configmap mcp-config \
   --from-literal=SQLITE_DB_DIR=/data \
-  --from-literal=PORT=3001
+  --from-literal=PORT=3001 \
+  -n kagent
 ```
 
 Create a file named `deployment.yaml`:
@@ -103,6 +123,7 @@ apiVersion: apps/v1
 kind: Deployment
 metadata:
   name: mcp-sqlite-vec
+  namespace: kagent
   labels:
     app: mcp-sqlite-vec
 spec:
@@ -151,6 +172,7 @@ apiVersion: v1
 kind: Service
 metadata:
   name: mcp-sqlite-vec
+  namespace: kagent
 spec:
   selector:
     app: mcp-sqlite-vec
@@ -167,15 +189,17 @@ kubectl apply -f service.yaml
 
 ## Use the MCP server in kagent
 
-In the kagent UI, click on `Tools` and then on `Manage tool servers`.
+You can configure the MCP server in kagent using either the web UI or YAML manifests.
+
+### Option 1: Using the kagent dashboard
 
 ![manage tool servers](/images/manage-tool-servers.png "Manage tool servers")
 
-Click on `Add Server`.
+In the kagent UI, click on the `Create` dropdown menu and select `New Tool Server`.
 
 Call it `sqlite-vec`.
 
-Select `URL` and use `http://mcp-sqlite-vec.default:3001/sse`.
+Select `URL` and use `http://mcp-sqlite-vec.kagent:3001/mcp`.
 
 ![add tool server](/images/add-tool-server.png "Add tool server")
 
@@ -185,9 +209,37 @@ After refreshing the page, you should see the `query-documentation` tool being d
 
 ![list tool servers](/images/list-tool-servers.png "List tool servers")
 
+### Option 2: Using YAML CRDs
+
+Create a `remote-mcpserver.yaml` file:
+
+```yaml
+apiVersion: kagent.dev/v1alpha2
+kind: RemoteMCPServer
+metadata:
+  name: live-demo
+  namespace: kagent
+status:
+spec:
+  description: ''
+  protocol: STREAMABLE_HTTP
+  sseReadTimeout: 5m0s
+  terminateOnClose: true
+  timeout: 5s
+  url: http://mcp-sqlite-vec.kagent:3001/mcp
+```
+
+Apply the ToolServer:
+
+```bash
+kubectl apply -f toolserver.yaml
+```
+
 ## Create the MCP agent
 
-Click on `New Agent`
+### Option 1: Using the kagent dashboard
+
+Click on the `Create` dropdown menu and select `New Agent`.
 
 Use the following information:
 
@@ -199,6 +251,42 @@ Click on `Add Tools`.
 
 Select the `query-documentation` tool.
 
+### Option 2: Using YAML CRDs
+
+Create an `agent.yaml` file:
+
+```yaml
+# Kagent Agent Configuration
+# This defines an AI agent that can answer questions about MCP documentation
+apiVersion: kagent.dev/v1alpha2
+kind: Agent
+metadata:
+  name: sqlite-vec
+  namespace: kagent
+spec:
+  declarative:
+    modelConfig: default-model-config
+    stream: true
+    systemMessage: ' Use your tool to answer any question about the Model Context Protocol (MCP). Use mcp for the product and latest for the version'
+    tools:
+      - mcpServer:
+          apiGroup: kagent.dev
+          kind: RemoteMCPServer
+          name: live-demo
+          toolNames:
+            - query_documentation
+        type: McpServer
+  description: |
+    The MCP agent is answering questions about MCP, using the MCP documentation
+  type: Declarative
+```
+
+Apply the Agent:
+
+```bash
+kubectl apply -f agent.yaml
+```
+
 ![list agents](/images/list-agents.png "List agents")
 
 Congratulations, the agent is ready!