Update Readme.md (#1719)

Author: Arsh-Kashyap

certified-connectors/CognitiveServicesQuestionAnswering/Readme.md

@@ -23,12 +23,29 @@ You will need the following to proceed:
 - A Microsoft Power Apps or Microsoft Power Automate plan with custom connector feature
 - The Power platform CLI tools
 
+## Building the connector for AAD Authentication 
+Since the APIs used by the connector are secured by Azure Active Directory (AD), we first need to set up a few things in Azure AD so that our connector can securely access the Microsoft Cognitive Services APIs.  After this setup, you can create and test the connector.
+
+### Set up an Azure AD application for your custom connector
+Since the connector uses OAuth as authentication type, we first need to register an application in Azure AD.  This application will be used to get the authorization token required to invoke rest APIs used by the connector on user's behalf.  You can read more about this [here](https://docs.microsoft.com/en-us/azure/active-directory/develop/authentication-scenarios) and follow the steps below:
+
+1. Create an Azure AD application
+This can be done using [Azure Portal](https://portal.azure.com), by following the steps [here](https://docs.microsoft.com/en-us/azure/active-directory/develop/quickstart-register-app).  Once created, note down the value of Application (Client) ID.  You will need this later.
+
+2. Configure (Update) your Azure AD application to access Microsoft Cognitive Services APIs.
+This step will ensure that your application can successfully retrieve an access token to Microsoft Cognitive Services rest APIs on behalf of your users.  To do this, follow the steps [here](https://docs.microsoft.com/en-us/azure/active-directory/develop/quickstart-configure-app-access-web-apis).
+    - For redirect URI, use "https://global.consent.azure-apim.net/redirect"
+    - For the credentials, use a client secret (and not certificates).  Remember to note the secret down, you will need this later and it is shown only once.
+    - For API permissions, make sure "Microsoft Cognitive Services" and "user_impersonation" are added.
+   
+At this point, we now have a valid Azure AD application that can be used to get permissions from end users and access Microsoft Cognitive Services. The next step for us is to create a custom connector.
+
 ## Deploying the connector
 
 Run the following command and follow the prompts:
 
 ```paconn
-paconn create --api-prop [Path to apiProperties.json] --api-def [Path to apiDefinition.swagger.json] --icon [Path to icon.png] 
+paconn create --api-prop [Path to apiProperties.json] --api-def [Path to apiDefinition.swagger.json] --icon [Path to icon.png] --secret <client_secret>
 ```
 
 ## Supported actions