[AI] [Agents] v1.1.0b2 (Azure#41325)

* [AI] [Agents] add is_valid_connection_id check to connection tool init

* Howie/async func b2 (Azure#41349)

* use as* `asyncio.gather` is used to make function tool calls in paral… (Azure#41311)

* use as* `asyncio.gather` is used to make function tool calls in parallel for `async` scenario.

* Fixed version

* update

* update version

* run black

* fix change_context usages (Azure#41362)

* fixing tracing for streaming when not using event handler (Azure#41424)

* fixing tracing for streaming when not using event handler

* fixing mypy warnings

* fixes

* changelog

* Update CHANGELOG.md

* Howie/samples with proj client (Azure#41426)

* Update samples with project client

* more samples

* update samples

* UPDATE

* update

* change tooling samples

* black and readme update

* update pip install

* Update changelog

* change  is_valid_connection_id to _is_valid_connection_id and run black

* fixed lint

* revert dev req

* Update README.md

* update

* fix readme

* added a new sample

* Update CHANGELOG.md with new release date

---------

Co-authored-by: Howie Leung <[email protected]>
Co-authored-by: Liudmila Molkova <[email protected]>
Co-authored-by: M-Hietala <[email protected]>

Author: 4 people

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

@@ -2,6 +2,18 @@
 
 # Release History
 
+## 1.1.0b2 (2025-06-09)
+
+### Bugs Fixed
+
+- `asyncio.gather` is used to make function tool calls in parallel for `async` scenario.
+- Adding instrumentation for create_thread_and_run.
+- Fixed a tracing related bug that caused process_thread_run span to not appear when streaming is used without event handler.
+
+### Sample updates
+- Changed all samples to use `AIProjectClient` which is recommended to specify endpoint and credential.
+- Added `sample_agents_stream_iteration_with_functions.py`
+
 ## 1.1.0b1 (2025-05-20)
 
 ### Features Added

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

@@ -124,12 +124,11 @@ Here is an example of how to create an Agent:
 <!-- SNIPPET:sample_agents_basics.create_agent -->
 
 ```python
-
-    agent = agents_client.create_agent(
-        model=os.environ["MODEL_DEPLOYMENT_NAME"],
-        name="my-agent",
-        instructions="You are helpful agent",
-    )
+agent = agents_client.create_agent(
+    model=os.environ["MODEL_DEPLOYMENT_NAME"],
+    name="my-agent",
+    instructions="You are helpful agent",
+)
 ```
 
 <!-- END SNIPPET -->
@@ -300,7 +299,8 @@ conn_id = os.environ["AZURE_BING_CONNECTION_ID"]
 bing = BingGroundingTool(connection_id=conn_id)
 
 # Create agent with the bing tool and process agent run
-with agents_client:
+with project_client:
+    agents_client = project_client.agents
     agent = agents_client.create_agent(
         model=os.environ["MODEL_DEPLOYMENT_NAME"],
         name="my-agent",
@@ -330,7 +330,14 @@ ai_search = AzureAISearchTool(
 )
 
 # Create agent with AI search tool and process agent run
-with agents_client:
+project_client = AIProjectClient(
+    endpoint=os.environ["PROJECT_ENDPOINT"],
+    credential=DefaultAzureCredential(),
+)
+
+with project_client:
+    agents_client = project_client.agents
+
     agent = agents_client.create_agent(
         model=os.environ["MODEL_DEPLOYMENT_NAME"],
         name="my-agent",
@@ -586,7 +593,7 @@ Below is an example of how to create an Azure Logic App utility tool and registe
 ```python
 
 # Create the agents client
-agents_client = AgentsClient(
+project_client = AIProjectClient(
     endpoint=os.environ["PROJECT_ENDPOINT"],
     credential=DefaultAzureCredential(),
 )
@@ -647,7 +654,9 @@ openapi_tool.add_definition(
 )
 
 # Create agent with OpenApi tool and process agent run
-with agents_client:
+with project_client:
+    agents_client = project_client.agents
+
     agent = agents_client.create_agent(
         model=os.environ["MODEL_DEPLOYMENT_NAME"],
         name="my-agent",
@@ -676,7 +685,9 @@ print(conn_id)
 fabric = FabricTool(connection_id=conn_id)
 
 # Create an Agent with the Fabric tool and process an Agent run
-with agents_client:
+with project_client:
+    agents_client = project_client.agents
+
     agent = agents_client.create_agent(
         model=os.environ["MODEL_DEPLOYMENT_NAME"],
         name="my-agent",
@@ -907,7 +918,7 @@ message = agents_client.messages.create(
 
 To process your message, you can use `runs.create`, `runs.create_and_process`, or `runs.stream`.
 
-`create_run` requests the Agent to process the message without polling for the result. If you are using `function tools` regardless as `toolset` or not, your code is responsible for polling for the result and acknowledging the status of `Run`. When the status is `requires_action`, your code is responsible for calling the function tools. For a code sample, visit [`sample_agents_functions.py`](https://github.com/Azure/azure-sdk-for-python/blob/main/sdk/ai/azure-ai-agents/samples/agents_tools/sample_agents_functions.py).
+`runs.create` requests the Agent to process the message without polling for the result. If you are using `function tools`, your code is responsible for polling for the result and acknowledging the status of `Run`. When the status is `requires_action`, your code is responsible for calling the function tools. For a code sample, visit [`sample_agents_functions.py`](https://github.com/Azure/azure-sdk-for-python/blob/main/sdk/ai/azure-ai-agents/samples/agents_tools/sample_agents_functions.py).
 
 Here is an example of `runs.create` and poll until the run is completed:
 
@@ -925,10 +936,20 @@ while run.status in ["queued", "in_progress", "requires_action"]:
 
 <!-- END SNIPPET -->
 
-To have the SDK poll on your behalf and call `function tools`, use the `create_and_process` method. Note that `function tools` will only be invoked if they are provided as `toolset` during the `create_agent` call.
+To have the SDK poll on your behalf and call `function tools`, supply your function implementations through `enable_auto_function_calls` along with `runs.create_and_process` method .
 
 Here is an example:
 
+```python
+functions = FunctionTool(user_functions)
+
+toolset = ToolSet()
+toolset.add(functions)
+
+# To enable tool calls executed automatically
+agents_client.enable_auto_function_calls(toolset)
+```
+
 <!-- SNIPPET:sample_agents_run_with_toolset.create_and_process_run -->
 
 ```python
@@ -937,9 +958,9 @@ run = agents_client.runs.create_and_process(thread_id=thread.id, agent_id=agent.
 
 <!-- END SNIPPET -->
 
-With streaming, polling need not be considered. If `function tools` are provided as `toolset` during the `create_agent` call, they will be invoked by the SDK.
+With streaming, polling need not be considered. If `function tools` were added to the agents, you should decide to have the function tools called manually or automatically.  Please visit [`manual function call sample`](https://github.com/Azure/azure-sdk-for-python/blob/main/sdk/ai/azure-ai-agents/samples/agents_streaming/sample_agents_stream_eventhandler_with_functions.py) or [`automatic function call sample`](https://github.com/Azure/azure-sdk-for-python/blob/main/sdk/ai/azure-ai-agents/samples/agents_streaming/sample_agents_stream_iteration_with_toolset.py).    
 
-Here is an example of streaming:
+Here is a basic example of streaming:
 
 <!-- SNIPPET:sample_agents_basics_stream_iteration.iterate_stream -->
 
@@ -1171,7 +1192,8 @@ scenario = os.path.basename(__file__)
 tracer = trace.get_tracer(__name__)
 
 with tracer.start_as_current_span(scenario):
-    with agents_client:
+    with project_client:
+        agents_client = project_client.agents
 ```
 
 <!-- END SNIPPET -->
@@ -1281,4 +1303,4 @@ additional questions or comments.
 [azure_sub]: https://azure.microsoft.com/free/
 [evaluators]: https://learn.microsoft.com/azure/ai-studio/how-to/develop/evaluate-sdk
 [azure_ai_evaluation]: https://learn.microsoft.com/python/api/overview/azure/ai-evaluation-readme
-[evaluator_library]: https://learn.microsoft.com/azure/ai-studio/how-to/evaluate-generative-ai-app#view-and-manage-the-evaluators-in-the-evaluator-library
+[evaluator_library]: https://learn.microsoft.com/azure/ai-studio/how-to/evaluate-generative-ai-app#view-and-manage-the-evaluators-in-the-evaluator-library

sdk/ai/azure-ai-agents/azure/ai/agents/_version.py

@@ -6,4 +6,4 @@
 # Changes may cause incorrect behavior and will be lost if the code is regenerated.
 # --------------------------------------------------------------------------
 
-VERSION = "1.1.0b1"
+VERSION = "1.1.0b2"

sdk/ai/azure-ai-agents/azure/ai/agents/models/_patch.py

@@ -852,15 +852,25 @@ def execute(self, tool_call: Any) -> Any:
 class ConnectionTool(Tool[ToolDefinitionT]):
     """
     A tool that requires connection ids.
-    Used as base class for Bing Grounding, Sharepoint, and Microsoft Fabric
+    Used as base class for Sharepoint and Microsoft Fabric
     """
 
     def __init__(self, connection_id: str):
         """
         Initialize ConnectionTool with a connection_id.
 
         :param connection_id: Connection ID used by tool. All connection tools allow only one connection.
+        :raises ValueError: If the connection id is invalid.
         """
+        if not _is_valid_connection_id(connection_id):
+            raise ValueError(
+                "Connection ID '"
+                + connection_id
+                + "' does not fit the format:"
+                + "'/subscriptions/<subscription_id>/resourceGroups/<resource_group_name>/"
+                + "providers/<provider_name>/accounts/<account_name>/projects/<project_name>/connections/<connection_name>'"
+            )
+
         self.connection_ids = [ToolConnection(connection_id=connection_id)]
 
     @property
@@ -890,10 +900,21 @@ def __init__(self, connection_id: str, market: str = "", set_lang: str = "", cou
         :param set_lang: The language to use for user interface strings when calling Bing API.
         :param count: The number of search results to return in the Bing API response.
         :param freshness: Filter search results by a specific time range.
-        
-        .. seealso:: 
+        :raises ValueError: If the connection id is invalid.
+
+        .. seealso::
            `Bing Web Search API Query Parameters <https://learn.microsoft.com/bing/search-apis/bing-web-search/reference/query-parameters>`_
         """
+
+        if not _is_valid_connection_id(connection_id):
+            raise ValueError(
+                "Connection ID '"
+                + connection_id
+                + "' does not fit the format:"
+                + "'/subscriptions/<subscription_id>/resourceGroups/<resource_group_name>/"
+                + "providers/<provider_name>/accounts/<account_name>/projects/<project_name>/connections/<connection_name>'"
+            )
+
         self._search_configurations = [
             BingGroundingSearchConfiguration(
                 connection_id=connection_id, market=market, set_lang=set_lang, count=count, freshness=freshness
@@ -1356,29 +1377,27 @@ def validate_tool_type(self, tool: Tool) -> None:
                 + "Please use AsyncFunctionTool instead and provide sync and/or async function(s)."
             )
 
+    async def _execute_single_tool_call(self, tool_call: Any):
+        try:
+            tool = self.get_tool(AsyncFunctionTool)
+            output = await tool.execute(tool_call)
+            return {"tool_call_id": tool_call.id, "output": str(output)}
+        except Exception as e:  # pylint: disable=broad-exception-caught
+            return {"tool_call_id": tool_call.id, "output": str(e)}
+
     async def execute_tool_calls(self, tool_calls: List[Any]) -> Any:
         """
-        Execute a tool of the specified type with the provided tool calls.
+        Execute a tool of the specified type with the provided tool calls concurrently.
 
         :param List[Any] tool_calls: A list of tool calls to execute.
         :return: The output of the tool operations.
         :rtype: Any
         """
-        tool_outputs = []
 
-        for tool_call in tool_calls:
-            try:
-                if tool_call.type == "function":
-                    tool = self.get_tool(AsyncFunctionTool)
-                    output = await tool.execute(tool_call)
-                    tool_output = {
-                        "tool_call_id": tool_call.id,
-                        "output": str(output),
-                    }
-                    tool_outputs.append(tool_output)
-            except Exception as e:  # pylint: disable=broad-exception-caught
-                tool_output = {"tool_call_id": tool_call.id, "output": str(e)}
-                tool_outputs.append(tool_output)
+        # Execute all tool calls concurrently
+        tool_outputs = await asyncio.gather(
+            *[self._execute_single_tool_call(tc) for tc in tool_calls if tc.type == "function"]
+        )
 
         return tool_outputs
 
@@ -1823,6 +1842,30 @@ def __exit__(self, exc_type, exc_val, exc_tb):
             close_method()
 
 
+def _is_valid_connection_id(connection_id: str) -> bool:
+    """
+    Validates if a string matches the Azure connection resource ID format.
+
+    The expected format is:
+    "/subscriptions/<AZURE_SUBSCRIPTION_ID>/resourceGroups/<RESOURCE_GROUP>/
+    providers/<AZURE_PROVIDER>/accounts/<ACCOUNT_NAME>/projects/<PROJECT_NAME>/
+    connections/<CONNECTION_NAME>"
+
+    :param connection_id: The connection ID string to validate
+    :type connection_id: str
+    :return: True if the string matches the expected format, False otherwise
+    :rtype: bool
+    """
+    pattern = (
+        r"^/subscriptions/[^/]+/resourceGroups/[^/]+/providers/[^/]+/accounts/[^/]+/projects/[^/]+/connections/[^/]+$"
+    )
+
+    # Check if the string matches the pattern
+    if re.match(pattern, connection_id):
+        return True
+    return False
+
+
 __all__: List[str] = [
     "AgentEventHandler",
     "AgentRunStream",