Merge branch 'main' into 438663-improve-nav-add-entities

Author: laujan

articles/ai-services/agents/how-to/connected-agents.md

@@ -6,7 +6,7 @@ services: cognitive-services
 manager: nitinme
 ms.service: azure
 ms.topic: how-to
-ms.date: 05/02/2025
+ms.date: 06/04/2025
 author: aahill
 ms.author: aahi
 recommendations: false
@@ -71,8 +71,14 @@ Checks the contract against internal standards or uploaded guidelines to identif
 
     "Review this document against company compliance guidelines and flag any deviations from the approved template."
 
+## Limitations
+
+* Connected agents cannot call local functions using the function calling tool. We recommend using the [OpenAPI tool](./tools/openapi-spec.md) or [Azure Functions](./tools/azure-functions.md) instead.
+* It is currently not possible to guarantee citations will be passed from connected agents. You can try using prompt engineering combined with different models to try and improve the possibility that citations will be outputted by the main agent, but results are subject to variability. 
+
 :::zone pivot="portal"
 
+
 ## Creating a multi-agent setup
 
 1. Navigate to the **Agents** page in the portal

articles/ai-services/agents/how-to/tools/bing-grounding.md

@@ -37,6 +37,13 @@ Grounding with Bing returns relevant search results to the customer's model depl
 
 The authorization will happen between Grounding with Bing Search service and Azure AI Foundry Agent Service. Any Bing search query that is generated and sent to Bing for the purposes of grounding is transferred, along with the resource key, outside of the Azure compliance boundary to the Grounding with Bing Search service. Grounding with Bing Search is subject to Bing's terms and do not have the same compliance standards and certifications as the Azure AI Foundry Agent Service, as described in the [Grounding with Bing Search Terms of Use](https://www.microsoft.com/bing/apis/grounding-legal). It is your responsibility to assess whether the use of Grounding with Bing Search in your agent meets your needs and requirements.
 
+## Supported capabilities and known issues
+- Grounding with Bing Search tool is designed to retrieve real-time information from web, NOT specific web domains.
+- NOT Recommended to **summarize** an entire web page.
+- Within one run, the AI model will evaluate the tool outputs and may decide to invoke the tool again for more information and context. AI model may also decide which piece(s) of tool outputs are used to generate the response.
+- Azure AI Agent service will return **AI model generated response** as output so end-to-end latency will be impacted pre-/post-processing of LLMs.
+- Grounding with Bing Search tool does NOT return the tool output to developers and end users.
+
 ## Usage support
 
 |Azure AI foundry support  | Python SDK |	C# SDK | JavaScript SDK | REST API |Basic agent setup | Standard agent setup |

articles/ai-services/agents/how-to/virtual-networks.md

@@ -54,6 +54,7 @@ For customers without an existing virtual network, the Standard Setup with Priva
     * `Microsoft.Search`
     * `Microsoft.Network`
     * `Microsoft.App`
+    * `Microsoft.ContainerService`
     * To use Bing Search tool: `Microsoft.Bing`
 
     ```console
@@ -104,6 +105,36 @@ For customers without an existing virtual network, the Standard Setup with Priva
         ```console
         az deployment group create --resource-group {my_resource_group} --template-file main-create.bicep
         ```
+
+1. Run the CheckCapabilityHostReadiness.ps1 and edit the command to add your subscription ID, resource group name, and your newly deployed AI Services account resource name.
+   
+   ```
+   .\CheckCapabilityHostReadiness.ps1 -subscriptionId "<your-sub-id>" -resourcegroup "<new-rg-name>" -accountname "<your-aiservices-name>"
+   ```
+   
+   If you don't want to run the PowerShell script, you can run a bash script instead, from the file CheckCapabilityHostReadiness.sh. Run the following two commands:
+   
+      ```
+      chmod +x CheckCapabilityHostReadiness.sh
+      ./CheckCapabilityHostReadiness.sh "<your-sub-id>" "<new-rg-name>" "<your-aiservices-name>"
+      ```
+      
+1. Deploy the main-project-caphost-create.bicep
+   
+   ```
+   az deployment group create --resource-group <new-rg-name> --template-file main-project-caphost-create.bicep
+   ```
+   
+   After running this script, you're required to input the following values:
+   
+   ```
+   Please provide string value for 'accountName' (? for help): <your-account-name>
+   Please provide string value for 'projectName' (? for help): <your-project-name>
+   Please provide string value for 'aiSearchName' (? for help): <your-search-name>
+   Please provide string value for 'azureStorageName' (? for help): <your-storage-name>
+   Please provide string value for 'cosmosDBName' (? for help): <your-cosmosdb-name>
+   ```
+
 For more details, see the [README](https://github.com/azure-ai-foundry/foundry-samples/tree/main/samples/microsoft/infrastructure-setup/15-private-network-standard-agent-setup).
 
 ## Deep Dive Standard Setup with Private Networking Template

articles/ai-services/agents/includes/quickstart-python.md

@@ -79,32 +79,32 @@ with project_client:
     )
     print(f"Created agent, ID: {agent.id}")
 
-# Create a thread for communication
-thread = project_client.agents.threads.create()
-print(f"Created thread, ID: {thread.id}")
-
-# Add a message to the thread
-message = project_client.agents.messages.create(
-    thread_id=thread.id,
-    role="user",  # Role of the message sender
-    content="What is the weather in Seattle today?",  # Message content
-)
-print(f"Created message, ID: {message['id']}")
-
-# Create and process an agent run
-run = project_client.agents.runs.create_and_process(thread_id=thread.id, agent_id=agent.id)
-print(f"Run finished with status: {run.status}")
-
-# Check if the run failed
-if run.status == "failed":
-    print(f"Run failed: {run.last_error}")
-
-# Fetch and log all messages
-messages = project_client.agents.messages.list(thread_id=thread.id)
-for message in messages.data:
-    print(f"Role: {message.role}, Content: {message.content}")
-
-# Delete the agent when done
-project_client.agents.delete_agent(agent.id)
-print("Deleted agent")
+    # Create a thread for communication
+    thread = project_client.agents.threads.create()
+    print(f"Created thread, ID: {thread.id}")
+    
+    # Add a message to the thread
+    message = project_client.agents.messages.create(
+        thread_id=thread.id,
+        role="user",  # Role of the message sender
+        content="What is the weather in Seattle today?",  # Message content
+    )
+    print(f"Created message, ID: {message['id']}")
+    
+    # Create and process an agent run
+    run = project_client.agents.runs.create_and_process(thread_id=thread.id, agent_id=agent.id)
+    print(f"Run finished with status: {run.status}")
+    
+    # Check if the run failed
+    if run.status == "failed":
+        print(f"Run failed: {run.last_error}")
+    
+    # Fetch and log all messages
+    messages = project_client.agents.messages.list(thread_id=thread.id)
+    for message in messages:
+        print(f"Role: {message.role}, Content: {message.content}")
+    
+    # Delete the agent when done
+    project_client.agents.delete_agent(agent.id)
+    print("Deleted agent")
 ```

articles/ai-services/language-service/conversational-language-understanding/concepts/best-practices.md

@@ -6,7 +6,7 @@ author: laujan
 manager: nitinme
 ms.service: azure-ai-language
 ms.topic: best-practice
-ms.date: 11/21/2024
+ms.date: 06/04/2025
 ms.author: lajanuar
 ms.custom: language-service-clu
 ---
@@ -112,7 +112,7 @@ If you enable this feature, the utterance count of your training set increases.
 
 ## Address model overconfidence
 
-Customers can use the LoraNorm traning configuration version if the model is being incorrectly overconfident. An example of this behavior can be like the following scenario where the model predicts the incorrect intent with 100% confidence. This score makes the confidence threshold project setting unusable.
+Customers can use the LoraNorm training configuration version if the model is being incorrectly overconfident. An example of this behavior can be like the following scenario where the model predicts the incorrect intent with 100% confidence. This score makes the confidence threshold project setting unusable.
 
 | Text |	Predicted intent |	Confidence score |
 |----|----|----|
@@ -272,6 +272,6 @@ After the request is sent, you can track the progress of the training job in Lan
 
 Caveats:
 
-- The None score threshold for the app (confidence threshold below which `topIntent` is marked as `None`) when you use this training configuration should be set to 0. This setting is used because this new training configuration attributes a certain portion of the in-domain probabilities to out of domain so that the model isn't incorrectly overconfident about in-domain utterances. As a result, users might see slightly reduced confidence scores for in-domain utterances as compared to the prod training configuration.
+- The None score threshold for the app (confidence threshold below which `topIntent` is marked as `None`) when you use this training configuration should be set to 0. This setting is used because this new training configuration attributes a certain portion of the in-domain probabilities to out of domain so that the model isn't incorrectly overconfident about in-domain utterances. As a result, users might see slightly reduced confidence scores for in-domain utterances as compared to the production training configuration.
 - We don't recommend this training configuration for apps with only two intents, such as `IntentA` and `None`, for example.
 - We don't recommend this training configuration for apps with a low number of utterances per intent. We highly recommend a minimum of 25 utterances per intent.

articles/ai-services/language-service/conversational-language-understanding/concepts/entity-components.md

@@ -6,7 +6,7 @@ author: laujan
 manager: nitinme
 ms.service: azure-ai-language
 ms.topic: conceptual
-ms.date: 11/21/2024
+ms.date: 06/04/2025
 ms.author: lajanuar
 ms.custom: language-service-clu
 ---

articles/ai-services/language-service/conversational-language-understanding/concepts/evaluation-metrics.md

@@ -6,7 +6,7 @@ author: laujan
 manager: nitinme
 ms.service: azure-ai-language
 ms.topic: conceptual
-ms.date: 11/21/2024
+ms.date: 06/04/2025
 ms.author: lajanuar
 ms.custom: language-service-clu
 ---

articles/ai-services/language-service/conversational-language-understanding/faq.md

@@ -6,7 +6,7 @@ author: laujan
 manager: nitinme
 ms.service: azure-ai-language
 ms.topic: faq
-ms.date: 11/21/2024
+ms.date: 06/04/2025
 ms.author: lajanuar
 ms.custom: mode-other
 ---

articles/ai-services/language-service/conversational-language-understanding/how-to/call-api.md

@@ -6,7 +6,7 @@ author: laujan
 manager: nitinme
 ms.service: azure-ai-language
 ms.topic: how-to
-ms.date: 11/21/2024
+ms.date: 06/04/2025
 ms.author: lajanuar
 ms.custom: language-service-clu
 ---
@@ -18,7 +18,7 @@ You can query the deployment programmatically through the [prediction API](https
 
 ## Test deployed model
 
-You can use Language Studio to submit an utterance, get predictions and visualize the results.
+You can use Language Studio to submit an utterance, get predictions, and visualize the results.
 
 [!INCLUDE [Test model](../includes/language-studio/test-model.md)]
 
@@ -32,7 +32,7 @@ You can use Language Studio to submit an utterance, get predictions and visualiz
 
 # [REST APIs](#tab/REST-APIs)
 
-First you will need to get your resource key and endpoint:
+First you'll need to get your resource key and endpoint:
 
 [!INCLUDE [Get keys and endpoint Azure Portal](../includes/get-keys-endpoint-azure.md)]
 
@@ -53,7 +53,7 @@ You can also use the client libraries provided by the Azure SDK to send requests
 
 1. Go to your resource overview page in the [Azure portal](https://portal.azure.com/#home)
 
-2. From the menu on the left side, select **Keys and Endpoint**. Use endpoint for the API requests and you will need the key for `Ocp-Apim-Subscription-Key` header.
+2. From the menu on the left side, select **Keys and Endpoint**. Use endpoint for the API requests and you'll need the key for `Ocp-Apim-Subscription-Key` header.
 
     :::image type="content" source="../../custom-text-classification/media/get-endpoint-azure.png" alt-text="A screenshot showing a key and endpoint in the Azure portal." lightbox="../../custom-text-classification/media/get-endpoint-azure.png":::
 

articles/ai-services/language-service/conversational-language-understanding/language-support.md

@@ -11,7 +11,7 @@ ms.author: lajanuar
 ms.custom: language-service-clu
 ---
 
-# Language support for conversational language understanding
+# Language support for Conversational Language Understanding (CLU)
 
 Use this article to learn about the languages currently supported by CLU feature.