refactor: Agent framework v2 changes.

Author: Avijit-Microsoft

.github/workflows/deploy-KMGeneric.yml

@@ -118,7 +118,7 @@ jobs:
           echo "Generated SOLUTION_PREFIX: ${UNIQUE_SOLUTION_PREFIX}" 
       - name: Determine Tag Name Based on Branch
         id: determine_tag
-        run: echo "tagname=${{ github.ref_name == 'main' && 'latest_waf' || github.ref_name == 'dev' && 'dev' || github.ref_name == 'demo' && 'demo' || github.ref_name == 'dependabotchanges' && 'dependabotchanges' || 'latest_waf' }}" >> $GITHUB_OUTPUT
+        run: echo "tagname=${{ github.ref_name == 'main' && 'latest_afv2' || github.ref_name == 'dev' && 'dev' || github.ref_name == 'demo' && 'demo' || github.ref_name == 'dependabotchanges' && 'dependabotchanges' || 'latest_afv2' }}" >> $GITHUB_OUTPUT
       - name: Deploy Bicep Template
         id: deploy
         run: |

.github/workflows/docker-build.yml

@@ -61,7 +61,7 @@ jobs:
         id: determine_tag
         run: |
          if [[ "${{ github.ref_name }}" == "main" ]]; then
-          echo "tagname=latest_waf" >> $GITHUB_OUTPUT
+          echo "tagname=latest_afv2" >> $GITHUB_OUTPUT
          elif [[ "${{ github.ref_name }}" == "dev" ]]; then
           echo "tagname=dev" >> $GITHUB_OUTPUT
          elif [[ "${{ github.ref_name }}" == "demo" ]]; then

.github/workflows/job-azure-deploy.yml

@@ -456,8 +456,8 @@ jobs:
             echo "Current branch: $BRANCH_NAME"
             
             if [[ "$BRANCH_NAME" == "main" ]]; then
-              IMAGE_TAG="latest_waf"
-              echo "Using main branch - image tag: latest_waf"
+              IMAGE_TAG="latest_afv2"
+              echo "Using main branch - image tag: latest_afv2"
             elif [[ "$BRANCH_NAME" == "dev" ]]; then
               IMAGE_TAG="dev"
               echo "Using dev branch - image tag: dev"
@@ -471,8 +471,8 @@ jobs:
               IMAGE_TAG="dependabotchanges"
               echo "Using dependabotchanges branch - image tag: dependabotchanges"
             else
-              IMAGE_TAG="latest_waf"
-              echo "Using default for branch '$BRANCH_NAME' - image tag: latest_waf"
+              IMAGE_TAG="latest_afv2"
+              echo "Using default for branch '$BRANCH_NAME' - image tag: latest_afv2"
             fi
             
             echo "Using existing Docker image tag: $IMAGE_TAG"

README.md

@@ -22,7 +22,7 @@ Analysts working with large volumes of conversational data can use this solution
 Solution overview
 </h2>
 
-Leverages Azure Content Understanding, Foundry IQ, Azure OpenAI Service, Semantic Kernel, Azure SQL Database, and Cosmos DB to process large volumes of conversational data. Audio and text inputs are analyzed through event-driven pipelines to extract and vectorize key information, orchestrate intelligent responses, and power an interactive web front-end for exploring insights using natural language.
+Leverages Azure Content Understanding, Foundry IQ, Azure OpenAI Service, Azure AI Agent Framework, Azure SQL Database, and Cosmos DB to process large volumes of conversational data. Audio and text inputs are analyzed through event-driven pipelines to extract and vectorize key information, orchestrate intelligent responses, and power an interactive web front-end for exploring insights using natural language.
 
 ### Solution architecture
 |![image](./documents/Images/ReadMe/solution-architecture.png)|
@@ -101,14 +101,13 @@ _Note: This is not meant to outline all costs as selected SKUs, scaled use, cust
 | [Microsoft Foundry](https://learn.microsoft.com/en-us/azure/ai-foundry) | Used to orchestrate and build AI workflows that combine Azure AI services. | Free Tier | [Pricing](https://azure.microsoft.com/pricing/details/ai-studio/) |
 | [Foundry IQ](https://learn.microsoft.com/en-us/azure/search/search-what-is-azure-search) | Powers vector-based semantic search for retrieving indexed conversation data. | Standard S1; costs scale with document count and replica/partition settings. | [Pricing](https://azure.microsoft.com/pricing/details/search/) |
 | [Azure Storage Account](https://learn.microsoft.com/en-us/azure/storage/common/storage-account-overview) | Stores transcripts, intermediate outputs, and application assets. | Standard LRS; usage-based cost by storage/operations. | [Pricing](https://azure.microsoft.com/pricing/details/storage/blobs/) |
-| [Azure Key Vault](https://learn.microsoft.com/en-us/azure/key-vault/general/overview) | Secures secrets, credentials, and keys used across the application. | Standard Tier; cost per operation (e.g., secret retrieval). | [Pricing](https://azure.microsoft.com/pricing/details/key-vault/) |
+
 | [Azure AI Services (OpenAI)](https://learn.microsoft.com/en-us/azure/cognitive-services/openai/overview) | Enables language understanding, summarization, entity extraction, and chat capabilities using GPT models. | S0 Tier; pricing depends on token volume and model used (e.g., GPT-4o-mini). | [Pricing](https://azure.microsoft.com/pricing/details/cognitive-services/) |
 | [Azure Container Apps](https://learn.microsoft.com/en-us/azure/container-apps/overview) | Hosts microservices and APIs powering the front-end and backend orchestration. | Consumption plan with 0.5 vCPU, 1GiB memory; includes a free usage tier. | [Pricing](https://azure.microsoft.com/pricing/details/container-apps/) |
 | [Azure Container Registry](https://learn.microsoft.com/en-us/azure/container-registry/container-registry-intro) | Stores and serves container images used by Azure Container Apps. | Basic Tier; fixed daily cost per registry. | [Pricing](https://azure.microsoft.com/pricing/details/container-registry/) |
 | [Azure Monitor / Log Analytics](https://learn.microsoft.com/en-us/azure/azure-monitor/logs/log-analytics-overview) | Collects and analyzes telemetry and logs from services and containers. | Pay-as-you-go; charges based on data ingestion volume. | [Pricing](https://azure.microsoft.com/pricing/details/monitor/) |
 | [Azure SQL Database](https://learn.microsoft.com/en-us/azure/azure-sql/database/sql-database-paas-overview) | Stores structured data including insights, metadata, and indexed results. | General Purpose Tier; can be provisioned or serverless. Fixed cost if provisioned. | [Pricing](https://azure.microsoft.com/pricing/details/azure-sql-database/single/) |
 | [Azure Cosmos DB](https://learn.microsoft.com/en-us/azure/cosmos-db/introduction) | Used for fast, globally distributed NoSQL data storage for chat history and vector metadata. | Autoscale or provisioned throughput; fixed minimum cost if provisioned. | [Pricing](https://azure.microsoft.com/en-us/pricing/details/cosmos-db/autoscale-provisioned/) |
-| [Azure Functions](https://learn.microsoft.com/en-us/azure/azure-functions/functions-overview) | Executes lightweight, serverless backend logic and event-driven workflows. | Consumption Tier; billed per execution and duration. | [Pricing](https://azure.microsoft.com/en-us/pricing/details/functions/) |
 
 
 <br/>
@@ -173,9 +172,7 @@ Supporting documentation
 
 ### Security guidelines
 
-This solution uses [Azure Key Vault](https://learn.microsoft.com/en-us/azure/key-vault/general/overview) to securely store secrets, connection strings, and API keys required by application components.
-
-It also leverages [Managed Identity](https://learn.microsoft.com/en-us/entra/identity/managed-identities-azure-resources/overview) for secure access to Azure resources during local development and production deployment, eliminating the need for hard-coded credentials.
+This solution leverages [Managed Identity](https://learn.microsoft.com/en-us/entra/identity/managed-identities-azure-resources/overview) for secure access to Azure resources during local development and production deployment, eliminating the need for hard-coded credentials.
 
 To maintain strong security practices, it is recommended that GitHub repositories built on this solution enable [GitHub secret scanning](https://docs.github.com/code-security/secret-scanning/about-secret-scanning) to detect accidental secret exposure.
 

azure.yaml

@@ -18,7 +18,10 @@ hooks:
       run: |
         Write-Host "Web app URL: "
         Write-Host "$env:WEB_APP_URL" -ForegroundColor Cyan
-        Write-Host "`nCreate and activate a virtual environment if not already done, then run the following command in your Bash terminal. It will grant the necessary permissions between resources and your user account, and also process and load the sample data into the application."
+
+        Write-Host "`nCreate and activate a virtual environment if not already done, then run the following command in the bash terminal to create agents:"
+        Write-Host "bash ./infra/scripts/run_create_agents_scripts.sh" -ForegroundColor Cyan
+        Write-Host "`nRun the following command in your Bash terminal. It will grant the necessary permissions between resources and your user account, and also process and load the sample data into the application."
         Write-Host "bash ./infra/scripts/process_sample_data.sh" -ForegroundColor Cyan
       shell: pwsh
       continueOnError: false
@@ -27,8 +30,11 @@ hooks:
       run: |
         echo "Web app URL: "
         echo $WEB_APP_URL
+
+        echo "\nCreate and activate a virtual environment if not already done, then run the following command in the bash terminal to create agents:"
+        echo "bash ./infra/scripts/run_create_agents_scripts.sh"
         echo ""
-        echo "Create and activate a virtual environment if not already done, then run the following command in your Bash terminal. It will grant the necessary permissions between resources and your user account, and also process and load the sample data into the application."
+        echo "\nRun the following command in your Bash terminal. It will grant the necessary permissions between resources and your user account, and also process and load the sample data into the application."
         echo "bash ./infra/scripts/process_sample_data.sh"
       shell: sh
       continueOnError: false

documents/AVMPostDeploymentGuide.md

@@ -58,9 +58,36 @@ cd Conversation-Knowledge-Mining-Solution-Accelerator
 
 ---
 
-### Step 2: Run the Data Processing Script
+### Step 2: Create and Activate Python Virtual Environment
 
-#### 2.1 Login to Azure
+#### 2.1 Create a Python Virtual Environment
+
+```shell
+python -m venv .venv
+```
+
+#### 2.2 Activate the Virtual Environment
+
+**For Windows (PowerShell):**
+```powershell
+.venv\Scripts\Activate.ps1
+```
+
+**For Windows (Bash):**
+```bash
+source .venv/Scripts/activate
+```
+
+**For Linux/macOS/VS Code Web (Bash):**
+```bash
+source .venv/bin/activate
+```
+
+---
+
+### Step 3: Create AI Agents
+
+#### 3.1 Login to Azure
 
 ```shell
 az login
@@ -71,7 +98,19 @@ az login
 > az login --use-device-code
 > ```
 
-#### 2.2 Execute the Script
+#### 3.2 Execute the Agent Creation Script
+
+Run the bash script from the output of the AVM deployment:
+
+```bash
+bash ./infra/scripts/run_create_agents_scripts.sh <Resource-Group-Name>
+```
+
+> ⚠️ **Important**: Replace `<Resource-Group-Name>` with your actual resource group name from the deployment.
+
+---
+
+### Step 4: Process Sample Data
 
 Run the bash script from the output of the AVM deployment:
 
@@ -83,7 +122,7 @@ bash ./infra/scripts/process_sample_data.sh <Resource-Group-Name>
 
 ---
 
-### Step 3: Access the Application
+### Step 5: Access the Application
 
 1. Navigate to the [Azure Portal](https://portal.azure.com)
 2. Open the **resource group** created during deployment
@@ -93,13 +132,13 @@ bash ./infra/scripts/process_sample_data.sh <Resource-Group-Name>
 
 ---
 
-### Step 4: Configure Authentication (Optional)
+### Step 6: Configure Authentication (Optional)
 
 If you want to enable authentication for your application, follow the [App Authentication Guide](./AppAuthentication.md).
 
 ---
 
-### Step 5: Verify Data Processing
+### Step 7: Verify Data Processing
 
 Confirm your deployment is working correctly:
 
@@ -111,7 +150,7 @@ Confirm your deployment is working correctly:
 
 ---
 
-### 6. Customize with Your Own Data (Optional)
+### Step 8: Customize with Your Own Data (Optional)
 
 To replace the sample data with your own conversational data, follow these steps:
 
@@ -149,7 +188,7 @@ bash ./infra/scripts/process_custom_data.sh \
   <AI-Search-Name> <Search-Endpoint> \
   <AI-Foundry-Resource-ID> <CU-Foundry-Resource-ID> \
   <OpenAI-Endpoint> <Embedding-Model> <Deployment-Model> \
-  <CU-Endpoint> <AI-Agent-Endpoint> <CU-API-Version>
+  <CU-Endpoint> <CU-API-Version> <AI-Agent-Endpoint> <Solution-Name>
 ```
 
 #### VM Access for WAF Deployments

documents/CustomizeData.md

@@ -31,7 +31,7 @@ If you would like to update the solution to leverage your own data please follow
       <AI-Search-Name> <Search-Endpoint> \
       <AI-Foundry-Resource-ID> <CU-Foundry-Resource-ID> \
       <OpenAI-Endpoint> <Embedding-Model> <Deployment-Model> \
-      <CU-Endpoint> <AI-Agent-Endpoint> <CU-API-Version>
+      <CU-Endpoint> <CU-API-Version> <AI-Agent-Endpoint> <Solution-Name>
     ```
 
 ## How to Login to VM Using Azure Bastion

documents/CustomizingAzdParameters.md

@@ -19,7 +19,7 @@ By default this template will use the environment name as the prefix to prevent
 | `AZURE_OPENAI_API_VERSION`            | string  | `2025-01-01-preview`     | Specifies the API version for Azure OpenAI.                                |
 | `AZURE_OPENAI_DEPLOYMENT_MODEL_CAPACITY` | integer | `30`                     | Sets the GPT model capacity.                                               |
 | `AZURE_OPENAI_EMBEDDING_MODEL`            | string  | `text-embedding-ada-002` | Sets the name of the embedding model to use.                               |
-| `AZURE_ENV_IMAGETAG`                      | string  | `latest_waf`        | Sets the image tag (`latest_waf`, `dev`, `hotfix`, etc.).   |
+| `AZURE_ENV_IMAGETAG`                      | string  | `latest_afv2`        | Sets the image tag (`latest_afv2`, `dev`, `hotfix`, etc.).   |
 | `AZURE_OPENAI_EMBEDDING_MODEL_CAPACITY`   | integer | `80`                     | Sets the capacity for the embedding model deployment.                      |
 | `AZURE_ENV_LOG_ANALYTICS_WORKSPACE_ID`    | string  | Guide to get your [Existing Workspace ID](/documents/re-use-log-analytics.md)            | Reuses an existing Log Analytics Workspace instead of creating a new one.  |
 | `USE_LOCAL_BUILD`    | string  | `false`           | Indicates whether to use a local container build for deployment.  |

documents/DeploymentGuide.md

@@ -345,7 +345,34 @@ az login
 az login --use-device-code
 ```
 
-**4. Run the sample data processing script:**
+**4. Run the create agent script:**
+
+The `azd up` deployment output includes a ready-to-use bash script command. Look for the script in the deployment output and run it:
+
+```bash
+bash ./infra/scripts/run_create_agents_scripts.sh
+```
+
+**If you don't have `azd env` configured**, you'll need to pass parameters manually. The parameters are grouped by service for clarity:
+
+```bash
+bash ./infra/scripts/run_create_agents_scripts.sh \
+   <resource-group> \
+   <project-endpoint> <solution-name> <gpt-model-name> \
+   <ai-foundry-resource-id> <api-app-name> \
+   <azure-ai-search-connection-name> <azure-ai-search-index>
+```
+
+**Parameter Descriptions:**
+- **Resource Group Parameters:** Azure resource group name
+- **AI Foundry Parameters:** AI Foundry project endpoint URL and resource ID
+- **Solution Parameters:** Solution deployment name
+- **AI Model Parameters:** Deployed GPT model name
+- **Application Parameters:** API application name
+- **Search Parameters:** Azure AI Search connection name and index name
+
+
+**5. Run the sample data processing script:**
 
 The `azd up` deployment output includes a ready-to-use bash script command. Look for the script in the deployment output and run it:
 
@@ -363,7 +390,7 @@ bash ./infra/scripts/process_sample_data.sh \
   <AI-Search-Name> <Search-Endpoint> \
   <AI-Foundry-Resource-ID> <CU-Foundry-Resource-ID> \
   <OpenAI-Endpoint> <Embedding-Model> <Deployment-Model> \
-  <CU-Endpoint> <AI-Agent-Endpoint> <CU-API-Version> <Use-Case>
+  <CU-Endpoint> <CU-API-Version> <AI-Agent-Endpoint> <Use-Case> <Solution-Name>
 ```
 
 **Parameter Descriptions:**
@@ -375,6 +402,7 @@ bash ./infra/scripts/process_sample_data.sh \
 - **OpenAI Parameters:** OpenAI endpoint, embedding model name, and deployment model name
 - **Content Understanding Parameters:** CU endpoint, AI agent endpoint, CU API version
 - **Use Case:** Either `telecom` or `IT_helpdesk`
+- **Solution Parameters:** Solution deployment name
 
 > **Note:** All parameter values are available in the Azure Portal by navigating to your deployed resources, or from the `azd env get-values` command output.
 

documents/LocalDevelopmentSetup.md

@@ -532,6 +532,12 @@ AZURE_EXISTING_AI_PROJECT_RESOURCE_ID=<ai-project-resource-id>
 AZURE_AI_AGENT_ENDPOINT=<ai-agent-endpoint>
 AZURE_AI_AGENT_MODEL_DEPLOYMENT_NAME=<agent-model-deployment>
 
+# Agent Framework v2 Configuration (Set by deployment)
+AI_FOUNDRY_RESOURCE_ID=<ai-foundry-resource-id>
+API_APP_NAME=<api-app-name>
+AGENT_NAME_CONVERSATION=<conversation-agent-name>
+AGENT_NAME_TITLE=<title-agent-name>
+
 # Azure AI Search Configuration
 AZURE_AI_SEARCH_ENDPOINT=<search-endpoint>
 AZURE_AI_SEARCH_INDEX=call_transcripts_index
@@ -573,6 +579,7 @@ REACT_APP_LAYOUT_CONFIG=<layout-config-json>
 > - Set `APP_ENV=dev` for local development. This enables Azure CLI authentication.
 > - Ensure you're logged in via `az login` before running the backend.
 > - Set `APP_ENV=prod` only when deploying to Azure App Service with Managed Identity.
+> - **Agent Framework v2 Variables**: The `AI_FOUNDRY_RESOURCE_ID` and `API_APP_NAME` are automatically set during `azd up`. The `AGENT_NAME_CONVERSATION` and `AGENT_NAME_TITLE` are populated when you run the `run_create_agents_scripts.sh` script (see Step 4.4 in [Deployment Guide](./DeploymentGuide.md)).
 
 ### 4.3. Install Backend API Dependencies