Updates in post deployment script and readme files

Author: Ragini-Microsoft

docs/README_LOCAL.md

@@ -6,7 +6,6 @@
     These variables are required:
     - `AZURE_OPENAI_RESOURCE`
     - `AZURE_OPENAI_MODEL`
-    - `AZURE_OPENAI_KEY`
 
     These variables are optional:
     - `AZURE_OPENAI_TEMPERATURE`
@@ -174,6 +173,9 @@ Note: settings starting with `AZURE_SEARCH` are only needed when using Azure Ope
 
 | App Setting | Value | Note |
 | --- | --- | ------------- |
+|AZURE_AI_AGENT_API_VERSION|2025-01-01-preview| API version when using the Azure Foundry agent on your data.|
+|AZURE_AI_AGENT_ENDPOINT||The endpoint of the Azure AI foundry project|
+|AZURE_AI_AGENT_MODEL_DEPLOYMENT_NAME||The name of the gpt model|
 |AZURE_SEARCH_SERVICE||The name of your Azure AI Search resource|
 |AZURE_SEARCH_INDEX||The name of your Azure AI Search Index|
 |AZURE_SEARCH_KEY||An **admin key** for your Azure AI Search resource|
@@ -193,7 +195,6 @@ Note: settings starting with `AZURE_SEARCH` are only needed when using Azure Ope
 |AZURE_OPENAI_MODEL||The name of your model deployment|
 |AZURE_OPENAI_ENDPOINT||The endpoint of your Azure OpenAI resource.|
 |AZURE_OPENAI_MODEL_NAME|gpt-35-turbo-16k|The name of the model|
-|AZURE_OPENAI_KEY||One of the API keys of your Azure OpenAI resource|
 |AZURE_OPENAI_TEMPERATURE|0|What sampling temperature to use, between 0 and 2. Higher values like 0.8 will make the output more random, while lower values like 0.2 will make it more focused and deterministic. A value of 0 is recommended when using your data.|
 |AZURE_OPENAI_TOP_P|1.0|An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass. We recommend setting this to 1.0 when using your data.|
 |AZURE_OPENAI_MAX_TOKENS|1000|The maximum number of tokens allowed for the generated answer.|
@@ -211,6 +212,7 @@ Note: settings starting with `AZURE_SEARCH` are only needed when using Azure Ope
 |UI_SHOW_SHARE_BUTTON|True|Share button (right-top)
 |SANITIZE_ANSWER|False|Whether to sanitize the answer from Azure OpenAI. Set to True to remove any HTML tags from the response.|
 |USE_PROMPTFLOW|False|Use existing Promptflow deployed endpoint. If set to `True` then both `PROMPTFLOW_ENDPOINT` and `PROMPTFLOW_API_KEY` also need to be set.|
+|USE_AI_FOUNDRY_SDK|False|Boolean flag to determine whether to use the AI Foundry SDK instead of the OpenAI SDK.|
 |PROMPTFLOW_ENDPOINT||URL of the deployed Promptflow endpoint e.g. https://pf-deployment-name.region.inference.ml.azure.com/score|
 |PROMPTFLOW_API_KEY||Auth key for deployed Promptflow endpoint. Note: only Key-based authentication is supported.|
 |PROMPTFLOW_RESPONSE_TIMEOUT|120|Timeout value in seconds for the Promptflow endpoint to respond.|

infra/deploy_ai_foundry.bicep

@@ -190,15 +190,6 @@ resource tenantIdEntry 'Microsoft.KeyVault/vaults/secrets@2021-11-01-preview' =
   }
 }
 
-
-resource azureOpenAIApiKeyEntry 'Microsoft.KeyVault/vaults/secrets@2021-11-01-preview' = {
-  parent: keyVault
-  name: 'AZURE-OPENAI-KEY'
-  properties: {
-    value: aiFoundry.listKeys().key1 //aiServices_m.listKeys().key1
-  }
-}
-
 resource azureOpenAIDeploymentModel 'Microsoft.KeyVault/vaults/secrets@2021-11-01-preview' = {
   parent: keyVault
   name: 'AZURE-OPEN-AI-DEPLOYMENT-MODEL'

infra/deploy_app_service.bicep

@@ -35,10 +35,6 @@ param AzureOpenAIModel string
 @description('Azure Open AI Endpoint')
 param AzureOpenAIEndpoint string = ''
 
-@description('Azure OpenAI Key')
-@secure()
-param AzureOpenAIKey string
-
 param azureOpenAIApiVersion string
 param azureOpenaiResource string = ''
 param USE_CHAT_HISTORY_ENABLED string = ''
@@ -235,10 +231,6 @@ resource Website 'Microsoft.Web/sites@2020-06-01' = {
           name: 'AZURE_OPENAI_ENDPOINT'
           value: AzureOpenAIEndpoint
         }
-        {
-          name: 'AZURE_OPENAI_KEY'
-          value: AzureOpenAIKey
-        }
         {
           name: 'AZURE_OPENAI_RESOURCE'
           value: azureOpenaiResource

infra/main.bicep

@@ -386,7 +386,6 @@ module appserviceModule 'deploy_app_service.bicep' = {
     AzureSearchKey: keyVault.getSecret('AZURE-SEARCH-KEY')
     AzureOpenAIEndpoint:aifoundry.outputs.aoaiEndpoint
     AzureOpenAIModel: gptModelName 
-    AzureOpenAIKey:keyVault.getSecret('AZURE-OPENAI-KEY')
     azureOpenAIApiVersion: azureOpenaiAPIVersion //'2024-02-15-preview'
     azureOpenaiResource:aifoundry.outputs.aiFoundryName
     aiFoundryProjectName: aifoundry.outputs.aiFoundryProjectName

infra/main.json

@@ -5,7 +5,7 @@
     "_generator": {
       "name": "bicep",
       "version": "0.34.44.8038",
-      "templateHash": "15193785708652143745"
+      "templateHash": "11594891016150517886"
     }
   },
   "parameters": {
@@ -602,7 +602,7 @@
             "_generator": {
               "name": "bicep",
               "version": "0.34.44.8038",
-              "templateHash": "4081811946489579439"
+              "templateHash": "6457856750858199317"
             }
           },
           "parameters": {
@@ -1059,17 +1059,6 @@
                 "value": "[subscription().tenantId]"
               }
             },
-            {
-              "type": "Microsoft.KeyVault/vaults/secrets",
-              "apiVersion": "2021-11-01-preview",
-              "name": "[format('{0}/{1}', parameters('keyVaultName'), 'AZURE-OPENAI-KEY')]",
-              "properties": {
-                "value": "[listKeys(resourceId('Microsoft.CognitiveServices/accounts', variables('aiFoundryName')), '2025-04-01-preview').key1]"
-              },
-              "dependsOn": [
-                "[resourceId('Microsoft.CognitiveServices/accounts', variables('aiFoundryName'))]"
-              ]
-            },
             {
               "type": "Microsoft.KeyVault/vaults/secrets",
               "apiVersion": "2021-11-01-preview",
@@ -1494,14 +1483,6 @@
           "AzureOpenAIModel": {
             "value": "[parameters('gptModelName')]"
           },
-          "AzureOpenAIKey": {
-            "reference": {
-              "keyVault": {
-                "id": "[extensionResourceId(format('/subscriptions/{0}/resourceGroups/{1}', subscription().subscriptionId, resourceGroup().name), 'Microsoft.KeyVault/vaults', reference(extensionResourceId(format('/subscriptions/{0}/resourceGroups/{1}', subscription().subscriptionId, resourceGroup().name), 'Microsoft.Resources/deployments', 'deploy_ai_foundry'), '2022-09-01').outputs.keyvaultName.value)]"
-              },
-              "secretName": "AZURE-OPENAI-KEY"
-            }
-          },
           "azureOpenAIApiVersion": {
             "value": "[parameters('azureOpenaiAPIVersion')]"
           },
@@ -1552,7 +1533,7 @@
             "_generator": {
               "name": "bicep",
               "version": "0.34.44.8038",
-              "templateHash": "12581930310509122324"
+              "templateHash": "9043383413973668437"
             }
           },
           "parameters": {
@@ -1617,12 +1598,6 @@
                 "description": "Azure Open AI Endpoint"
               }
             },
-            "AzureOpenAIKey": {
-              "type": "securestring",
-              "metadata": {
-                "description": "Azure OpenAI Key"
-              }
-            },
             "azureOpenAIApiVersion": {
               "type": "string"
             },
@@ -1941,10 +1916,6 @@
                       "name": "AZURE_OPENAI_ENDPOINT",
                       "value": "[parameters('AzureOpenAIEndpoint')]"
                     },
-                    {
-                      "name": "AZURE_OPENAI_KEY",
-                      "value": "[parameters('AzureOpenAIKey')]"
-                    },
                     {
                       "name": "AZURE_OPENAI_RESOURCE",
                       "value": "[parameters('azureOpenaiResource')]"

infra/scripts/index_scripts/02_process_data.py

@@ -9,6 +9,8 @@
 from azure.search.documents import SearchClient
 from azure.storage.filedatalake import DataLakeServiceClient
 from azure.search.documents.indexes import SearchIndexClient
+from azure.identity import (DefaultAzureCredential,
+                                get_bearer_token_provider)
 
 
 key_vault_name = 'kv_to-be-replaced'
@@ -29,23 +31,24 @@ def get_secrets_from_kv(kv_name, secret_name):
 
 search_endpoint = get_secrets_from_kv(key_vault_name, "AZURE-SEARCH-ENDPOINT")
 search_key = get_secrets_from_kv(key_vault_name, "AZURE-SEARCH-KEY")
-openai_api_key = get_secrets_from_kv(key_vault_name, "AZURE-OPENAI-KEY")
 openai_api_base = get_secrets_from_kv(key_vault_name, "AZURE-OPENAI-ENDPOINT")
 openai_api_version = get_secrets_from_kv(key_vault_name, "AZURE-OPENAI-PREVIEW-API-VERSION")
 deployment = get_secrets_from_kv(key_vault_name, "AZURE-OPEN-AI-DEPLOYMENT-MODEL")
 
 
 # Function: Get Embeddings
-def get_embeddings(text: str, openai_api_base, openai_api_version, openai_api_key):
+def get_embeddings(text: str, openai_api_base, openai_api_version):
     model_id = "text-embedding-ada-002"
+    ad_token_provider = get_bearer_token_provider(
+            DefaultAzureCredential(), "https://cognitiveservices.azure.com/.default"
+        )
     client = AzureOpenAI(
         api_version=openai_api_version,
         azure_endpoint=openai_api_base,
-        api_key=openai_api_key
+        azure_ad_token_provider=ad_token_provider
     )
 
     embedding = client.embeddings.create(input=text, model=model_id).data[0].embedding
-
     return embedding
 
 
@@ -123,12 +126,12 @@ def prepare_search_doc(content, document_id):
         chunk_id = document_id + '_' + str(chunk_num).zfill(2)
 
         try:
-            v_contentVector = get_embeddings(str(chunk), openai_api_base, openai_api_version, openai_api_key)
+            v_contentVector = get_embeddings(str(chunk), openai_api_base, openai_api_version)
         except Exception as e:
             print(f"Error occurred: {e}. Retrying after 30 seconds...")
             time.sleep(30)
             try:
-                v_contentVector = get_embeddings(str(chunk), openai_api_base, openai_api_version, openai_api_key)
+                v_contentVector = get_embeddings(str(chunk), openai_api_base, openai_api_version)
             except Exception as e:
                 print(f"Retry failed: {e}. Setting v_contentVector to an empty list.")
                 v_contentVector = []

scripts/SAMPLE_DATA.md

@@ -9,7 +9,7 @@
 ## Create the config.json and .env
 - Within the scripts folder, create a new .env file. 
 - Copy and paste the contents from the scripts/.env.sample file. 
-- Replace the values for `<AZURE_OPENAI_RESOURCE>` and `<AZURE_OPENAI_KEY>` with the name of the Azure OpenAI resource and either KEY 1 or KEY 2.
+- Replace the value for `<AZURE_OPENAI_RESOURCE>` with the name of the Azure OpenAI resource and either KEY 1 or KEY 2.
 - Save the .env file.
 - Within the scripts folder, create a config file `config.json`. The format will be a list of JSON objects, with each object specifying a configuration of local data path and target search service and index. Assuming you used "Deploy to Azure" to deploy this solution accelerator, these values can be found within the resources themselves. If you did not change the Search Index name, the default value is: promissory-notes-index. Copy and paste the following script block into the config.json file and update accordingly. 
 

src/.env.sample

@@ -1,8 +1,10 @@
 # Chat
 DEBUG=True
+AZURE_AI_AGENT_API_VERSION=
+AZURE_AI_AGENT_ENDPOINT=
+AZURE_AI_AGENT_MODEL_DEPLOYMENT_NAME=
 AZURE_OPENAI_RESOURCE=
 AZURE_OPENAI_MODEL=
-AZURE_OPENAI_KEY=
 AZURE_OPENAI_MODEL_NAME=gpt-35-turbo-16k
 AZURE_OPENAI_TEMPERATURE=0
 AZURE_OPENAI_TOP_P=1.0