Add get_thread_run span, add missing ids to end run, update tracing samples to use toolset (Azure#40998)

Author: lmolkova

sdk/ai/azure-ai-agents/README.md

@@ -418,7 +418,7 @@ agent = await agents_client.create_agent(
 
 <!-- END SNIPPET -->
 
-Notice that if `enable_auto_function_calls` is called, the SDK will invoke the functions automatically during `create_and_process` or streaming.  If you prefer to execute them manually, refer to [`sample_agents_stream_eventhandler_with_functions.py`](https://github.com/Azure/azure-sdk-for-python/blob/main/sdk/ai/azure-ai-projects/samples/agents/sample_agents_stream_eventhandler_with_functions.py) or 
+Notice that if `enable_auto_function_calls` is called, the SDK will invoke the functions automatically during `create_and_process` or streaming.  If you prefer to execute them manually, refer to [`sample_agents_stream_eventhandler_with_functions.py`](https://github.com/Azure/azure-sdk-for-python/blob/main/sdk/ai/azure-ai-projects/samples/agents/sample_agents_stream_eventhandler_with_functions.py) or
 [`sample_agents_functions.py`](https://github.com/Azure/azure-sdk-for-python/blob/main/sdk/ai/azure-ai-projects/samples/agents/sample_agents_functions.py)
 
 ### Create Agent With Azure Function Call
@@ -1169,7 +1169,7 @@ from opentelemetry import trace
 from azure.monitor.opentelemetry import configure_azure_monitor
 
 # Enable Azure Monitor tracing
-application_insights_connection_string = os.environ["AI_APPINSIGHTS_CONNECTION_STRING"]
+application_insights_connection_string = os.environ["APPLICATIONINSIGHTS_CONNECTION_STRING"]
 configure_azure_monitor(connection_string=application_insights_connection_string)
 
 # enable additional instrumentations

sdk/ai/azure-ai-agents/azure/ai/agents/telemetry/_ai_agents_instrumentor.py

@@ -61,6 +61,7 @@ class OperationName(Enum):
     CREATE_THREAD = "create_thread"
     CREATE_MESSAGE = "create_message"
     START_THREAD_RUN = "start_thread_run"
+    GET_THREAD_RUN = "get_thread_run"
     EXECUTE_TOOL = "execute_tool"
     LIST_MESSAGES = "list_messages"
     LIST_RUN_STEPS = "list_run_steps"

sdk/ai/azure-ai-agents/azure/ai/agents/telemetry/_utils.py

@@ -18,10 +18,10 @@ def __init__(self, agents_client: AgentsClient):
         self.agents_client = agents_client
 
     def enable_azure_monitor_tracing(self):
-        application_insights_connection_string = os.environ.get("AI_APPINSIGHTS_CONNECTION_STRING")
+        application_insights_connection_string = os.environ.get("APPLICATIONINSIGHTS_CONNECTION_STRING")
         if not application_insights_connection_string:
-            print("AI_APPINSIGHTS_CONNECTION_STRING environment variable was not set.")
-            print("Please create AI_APPINSIGHTS_CONNECTION_STRING with the Application Insights,")
+            print("APPLICATIONINSIGHTS_CONNECTION_STRING environment variable was not set.")
+            print("Please create APPLICATIONINSIGHTS_CONNECTION_STRING with the Application Insights,")
             print("connection string. It should be enabled for this project.")
             print("Enable it via the 'Tracing' tab in your AI Foundry project page.")
             exit()

sdk/ai/azure-ai-agents/samples/agents_multiagent/utils/agent_trace_configurator.py

@@ -22,7 +22,7 @@
       messages, which may contain personal data. False by default.
     * AZURE_TRACING_GEN_AI_CONTENT_RECORDING_ENABLED - Optional. Set to `true` to trace the content of chat
       messages, which may contain personal data. False by default.
-    * AI_APPINSIGHTS_CONNECTION_STRING - Set to the connection string of your Application Insights resource.
+    * APPLICATIONINSIGHTS_CONNECTION_STRING - Set to the connection string of your Application Insights resource.
       This is used to send telemetry data to Azure Monitor. You can also get the connection string programmatically
       from AIProjectClient using the `telemetry.get_connection_string` method. A code sample showing how to do this
       can be found in the `sample_telemetry_async.py` file in the azure-ai-projects telemetry samples.
@@ -49,7 +49,7 @@ async def main() -> None:
         )
 
         # Enable Azure Monitor tracing
-        application_insights_connection_string = os.environ["AI_APPINSIGHTS_CONNECTION_STRING"]
+        application_insights_connection_string = os.environ["APPLICATIONINSIGHTS_CONNECTION_STRING"]
         configure_azure_monitor(connection_string=application_insights_connection_string)
 
         # enable additional instrumentations
@@ -74,16 +74,7 @@ async def main() -> None:
                 )
                 print(f"Created message, message ID: {message.id}")
 
-                run = await agents_client.runs.create(thread_id=thread.id, agent_id=agent.id)
-
-                # Poll the run as long as run status is queued or in progress
-                while run.status in ["queued", "in_progress", "requires_action"]:
-                    # Wait for a second
-                    time.sleep(1)
-                    run = await agents_client.runs.get(thread_id=thread.id, run_id=run.id)
-
-                    print(f"Run status: {run.status}")
-
+                run = await agents_client.runs.create_and_process(thread_id=thread.id, agent_id=agent.id)
                 print(f"Run completed with status: {run.status}")
 
                 await agents_client.delete_agent(agent.id)

sdk/ai/azure-ai-agents/samples/agents_telemetry/sample_agents_basics_async_with_azure_monitor_tracing.py

@@ -76,16 +76,7 @@ async def main() -> None:
             )
             print(f"Created message, message ID: {message.id}")
 
-            run = await agent_client.runs.create(thread_id=thread.id, agent_id=agent.id)
-
-            # Poll the run as long as run status is queued or in progress
-            while run.status in ["queued", "in_progress", "requires_action"]:
-                # Wait for a second
-                time.sleep(1)
-                run = await agent_client.runs.get(thread_id=thread.id, run_id=run.id)
-
-                print(f"Run status: {run.status}")
-
+            run = await agent_client.runs.create_and_process(thread_id=thread.id, agent_id=agent.id)
             print(f"Run completed with status: {run.status}")
 
             await agent_client.delete_agent(agent.id)

sdk/ai/azure-ai-agents/samples/agents_telemetry/sample_agents_basics_async_with_console_tracing.py

@@ -22,7 +22,7 @@
        the "Models + endpoints" tab in your Azure AI Foundry project.
     3) AZURE_TRACING_GEN_AI_CONTENT_RECORDING_ENABLED - Optional. Set to `true` to trace the content of chat
        messages, which may contain personal data. False by default.
-    4) AI_APPINSIGHTS_CONNECTION_STRING - Set to the connection string of your Application Insights resource.
+    4) APPLICATIONINSIGHTS_CONNECTION_STRING - Set to the connection string of your Application Insights resource.
        This is used to send telemetry data to Azure Monitor. You can also get the connection string programmatically
        from AIProjectClient using the `telemetry.get_connection_string` method. A code sample showing how to do this
        can be found in the `sample_telemetry.py` file in the azure-ai-projects telemetry samples.
@@ -43,7 +43,7 @@
 from azure.monitor.opentelemetry import configure_azure_monitor
 
 # Enable Azure Monitor tracing
-application_insights_connection_string = os.environ["AI_APPINSIGHTS_CONNECTION_STRING"]
+application_insights_connection_string = os.environ["APPLICATIONINSIGHTS_CONNECTION_STRING"]
 configure_azure_monitor(connection_string=application_insights_connection_string)
 
 # enable additional instrumentations
@@ -70,15 +70,8 @@
         )
         print(f"Created message, message ID: {message.id}")
 
-        run = agents_client.runs.create(thread_id=thread.id, agent_id=agent.id)
-
-        # Poll the run as long as run status is queued or in progress
-        while run.status in ["queued", "in_progress", "requires_action"]:
-            # Wait for a second
-            time.sleep(1)
-            run = agents_client.runs.get(thread_id=thread.id, run_id=run.id)
-
-            print(f"Run status: {run.status}")
+        run = agents_client.runs.create_and_process(thread_id=thread.id, agent_id=agent.id)
+        print(f"Run completed with status: {run.status}")
 
         agents_client.delete_agent(agent.id)
         print("Deleted agent")

sdk/ai/azure-ai-agents/samples/agents_telemetry/sample_agents_basics_with_azure_monitor_tracing.py

@@ -23,7 +23,7 @@
 
     Set these environment variables with your own values:
     1) PROJECT_ENDPOINT - the Azure AI Agents endpoint.
-    2) MODEL_DEPLOYMENT_NAME - The deployment name of the AI model, as found under the "Name" column in 
+    2) MODEL_DEPLOYMENT_NAME - The deployment name of the AI model, as found under the "Name" column in
        the "Models + endpoints" tab in your Azure AI Foundry project.
     3) AZURE_TRACING_GEN_AI_CONTENT_RECORDING_ENABLED - Optional. Set to `true` to trace the content of chat
        messages, which may contain personal data. False by default.
@@ -71,15 +71,8 @@
         message = agents_client.messages.create(thread_id=thread.id, role="user", content="Hello, tell me a joke")
         print(f"Created message, message ID: {message.id}")
 
-        run = agents_client.runs.create(thread_id=thread.id, agent_id=agent.id)
-
-        # Poll the run as long as run status is queued or in progress
-        while run.status in ["queued", "in_progress", "requires_action"]:
-            # Wait for a second
-            time.sleep(1)
-            run = agents_client.runs.get(thread_id=thread.id, run_id=run.id)
-
-            print(f"Run status: {run.status}")
+        run = agents_client.runs.create_and_process(thread_id=thread.id, agent_id=agent.id)
+        print(f"Run completed with status: {run.status}")
 
         agents_client.delete_agent(agent.id)
         print("Deleted agent")

sdk/ai/azure-ai-agents/samples/agents_telemetry/sample_agents_basics_with_console_tracing.py

@@ -24,7 +24,7 @@
 
     Set these environment variables with your own values:
     1) PROJECT_ENDPOINT - the Azure AI Agents endpoint.
-    2) MODEL_DEPLOYMENT_NAME - The deployment name of the AI model, as found under the "Name" column in 
+    2) MODEL_DEPLOYMENT_NAME - The deployment name of the AI model, as found under the "Name" column in
        the "Models + endpoints" tab in your Azure AI Foundry project.
     3) AZURE_TRACING_GEN_AI_CONTENT_RECORDING_ENABLED - Optional. Set to `true` to trace the content of chat
        messages, which may contain personal data. False by default.
@@ -99,15 +99,8 @@ def on_end(self, span: ReadableSpan):
         message = agents_client.messages.create(thread_id=thread.id, role="user", content="Hello, tell me a joke")
         print(f"Created message, message ID: {message.id}")
 
-        run = agents_client.runs.create(thread_id=thread.id, agent_id=agent.id)
-
-        # Poll the run as long as run status is queued or in progress
-        while run.status in ["queued", "in_progress", "requires_action"]:
-            # Wait for a second
-            time.sleep(1)
-            run = agents_client.runs.get(thread_id=thread.id, run_id=run.id)
-
-            print(f"Run status: {run.status}")
+        run = agents_client.runs.create_and_process(thread_id=thread.id, agent_id=agent.id)
+        print(f"Run completed with status: {run.status}")
 
         agents_client.delete_agent(agent.id)
         print("Deleted agent")

sdk/ai/azure-ai-agents/samples/agents_telemetry/sample_agents_basics_with_console_tracing_custom_attributes.py

@@ -22,7 +22,7 @@
        the "Models + endpoints" tab in your Azure AI Foundry project.
     3) AZURE_TRACING_GEN_AI_CONTENT_RECORDING_ENABLED - Optional. Set to `true` to trace the content of chat
        messages, which may contain personal data. False by default.
-    4) AI_APPINSIGHTS_CONNECTION_STRING - Set to the connection string of your Application Insights resource.
+    4) APPLICATIONINSIGHTS_CONNECTION_STRING - Set to the connection string of your Application Insights resource.
        This is used to send telemetry data to Azure Monitor. You can also get the connection string programmatically
        from AIProjectClient using the `telemetry.get_connection_string` method. A code sample showing how to do this
        can be found in the `sample_telemetry.py` file in the azure-ai-projects telemetry samples.
@@ -79,7 +79,7 @@ def on_unhandled_event(self, event_type: str, event_data: Any) -> None:
 
 
 # Enable Azure Monitor tracing
-application_insights_connection_string = os.environ["AI_APPINSIGHTS_CONNECTION_STRING"]
+application_insights_connection_string = os.environ["APPLICATIONINSIGHTS_CONNECTION_STRING"]
 configure_azure_monitor(connection_string=application_insights_connection_string)
 
 scenario = os.path.basename(__file__)

sdk/ai/azure-ai-agents/samples/agents_telemetry/sample_agents_stream_eventhandler_with_azure_monitor_tracing.py

@@ -5,12 +5,12 @@
 
 """
 DESCRIPTION:
-    This sample demonstrates how to use basic agent operations with function tools from
+    This sample demonstrates how to use agent operations with toolset from
     the Azure Agents service using a synchronous client with Azure Monitor tracing.
     View the results in the "Tracing" tab in your Azure AI Foundry project page.
 
 USAGE:
-    python sample_agents_functions_with_azure_monitor_tracing.py
+    python sample_agents_toolset_with_azure_monitor_tracing.py
 
     Before running the sample:
 
@@ -22,7 +22,7 @@
        the "Models + endpoints" tab in your Azure AI Foundry project.
     3) AZURE_TRACING_GEN_AI_CONTENT_RECORDING_ENABLED - Optional. Set to `true` to trace the content of chat
        messages, which may contain personal data. False by default.
-    4) AI_APPINSIGHTS_CONNECTION_STRING - Set to the connection string of your Application Insights resource.
+    4) APPLICATIONINSIGHTS_CONNECTION_STRING - Set to the connection string of your Application Insights resource.
        This is used to send telemetry data to Azure Monitor. You can also get the connection string programmatically
        from AIProjectClient using the `telemetry.get_connection_string` method. A code sample showing how to do this
        can be found in the `sample_telemetry.py` file in the azure-ai-projects telemetry samples.
@@ -34,9 +34,7 @@
 from azure.identity import DefaultAzureCredential
 from azure.ai.agents.models import (
     FunctionTool,
-    RequiredFunctionToolCall,
-    SubmitToolOutputsAction,
-    ToolOutput,
+    ToolSet,
     ListSortOrder,
 )
 from opentelemetry import trace
@@ -49,7 +47,7 @@
 )
 
 # Enable Azure Monitor tracing
-application_insights_connection_string = os.environ["AI_APPINSIGHTS_CONNECTION_STRING"]
+application_insights_connection_string = os.environ["APPLICATIONINSIGHTS_CONNECTION_STRING"]
 configure_azure_monitor(connection_string=application_insights_connection_string)
 
 # enable additional instrumentations if needed
@@ -90,6 +88,11 @@ def fetch_weather(location: str) -> str:
 
 # Initialize function tool with user function
 functions = FunctionTool(functions=user_functions)
+toolset = ToolSet()
+toolset.add(functions)
+
+# To enable tool calls executed automatically
+agents_client.enable_auto_function_calls(toolset)
 
 with tracer.start_as_current_span(scenario):
     with agents_client:
@@ -98,7 +101,7 @@ def fetch_weather(location: str) -> str:
             model=os.environ["MODEL_DEPLOYMENT_NAME"],
             name="my-agent",
             instructions="You are a helpful agent",
-            tools=functions.definitions,
+            toolset=toolset
         )
         print(f"Created agent, ID: {agent.id}")
 
@@ -112,42 +115,7 @@ def fetch_weather(location: str) -> str:
         )
         print(f"Created message, ID: {message.id}")
 
-        run = agents_client.runs.create(thread_id=thread.id, agent_id=agent.id)
-        print(f"Created run, ID: {run.id}")
-
-        while run.status in ["queued", "in_progress", "requires_action"]:
-            time.sleep(1)
-            run = agents_client.runs.get(thread_id=thread.id, run_id=run.id)
-
-            if run.status == "requires_action" and isinstance(run.required_action, SubmitToolOutputsAction):
-                tool_calls = run.required_action.submit_tool_outputs.tool_calls
-                if not tool_calls:
-                    print("No tool calls provided - cancelling run")
-                    agents_client.runs.cancel(thread_id=thread.id, run_id=run.id)
-                    break
-
-                tool_outputs = []
-                for tool_call in tool_calls:
-                    if isinstance(tool_call, RequiredFunctionToolCall):
-                        try:
-                            output = functions.execute(tool_call)
-                            tool_outputs.append(
-                                ToolOutput(
-                                    tool_call_id=tool_call.id,
-                                    output=output,
-                                )
-                            )
-                        except Exception as e:
-                            print(f"Error executing tool_call {tool_call.id}: {e}")
-
-                print(f"Tool outputs: {tool_outputs}")
-                if tool_outputs:
-                    agents_client.runs.submit_tool_outputs(
-                        thread_id=thread.id, run_id=run.id, tool_outputs=tool_outputs
-                    )
-
-            print(f"Current run status: {run.status}")
-
+        run = agents_client.runs.create_and_process(thread_id=thread.id, agent_id=agent.id, toolset=toolset)
         print(f"Run completed with status: {run.status}")
 
         # Delete the agent when done