Merge branch 'main' into dev/dotnet_workflow/remove_timeout

.github/workflows/python-check-coverage.py

@@ -34,15 +34,14 @@ from dataclasses import dataclass
 #   (e.g., "packages/core/agent_framework/observability.py")
 # =============================================================================
 ENFORCED_TARGETS: set[str] = {
-    # Packages (sorted alphabetically)
-    "packages.anthropic.agent_framework_anthropic",
-    "packages.azure-ai-search.agent_framework_azure_ai_search",
+    # Packages
     "packages.azure-ai.agent_framework_azure_ai",
     "packages.core.agent_framework",
     "packages.core.agent_framework._workflows",
-    "packages.foundry.agent_framework_foundry",
-    "packages.openai.agent_framework_openai",
     "packages.purview.agent_framework_purview",
+    "packages.anthropic.agent_framework_anthropic",
+    "packages.azure-ai-search.agent_framework_azure_ai_search",
+    "packages.openai.agent_framework_openai",
     # Individual files (if you want to enforce specific files instead of whole packages)
     "packages/core/agent_framework/observability.py",
     # Add more targets here as coverage improves

dotnet/src/Microsoft.Agents.AI.Workflows/Execution/StreamingRunEventStream.cs

@@ -23,7 +23,8 @@ internal sealed class StreamingRunEventStream : IRunEventStream
     private readonly CancellationTokenSource _runLoopCancellation;
     private readonly bool _disableRunLoop;
     private Task? _runLoopTask;
-    private RunStatus _runStatus = RunStatus.NotStarted;
+    private volatile RunStatus _runStatus = RunStatus.NotStarted;
+
     private int _completionEpoch; // Tracks which completion signal belongs to which consumer iteration
 
     public StreamingRunEventStream(ISuperStepRunner stepRunner, bool disableRunLoop = false)
@@ -127,7 +128,7 @@ internal sealed class StreamingRunEventStream : IRunEventStream
 
                 // Wait for next input from the consumer
                 // Works for both Idle (no work) and PendingRequests (waiting for responses)
-                await this._inputWaiter.WaitForInputAsync(TimeSpan.FromSeconds(1), linkedSource.Token).ConfigureAwait(false);
+                await this._inputWaiter.WaitForInputAsync(linkedSource.Token).ConfigureAwait(false);
 
                 // When signaled, resume running
                 this._runStatus = RunStatus.Running;
@@ -209,7 +210,10 @@ internal sealed class StreamingRunEventStream : IRunEventStream
         [EnumeratorCancellation] CancellationToken cancellationToken = default)
     {
         // Get the current epoch - we'll only respond to completion signals from this epoch or later
-        int myEpoch = Volatile.Read(ref this._completionEpoch) + 1;
+        int currentEpoch = Volatile.Read(ref this._completionEpoch);
+
+        bool expectingFreshWork = this._stepRunner.HasUnprocessedMessages || this._runStatus == RunStatus.Running;
+        int myEpoch = expectingFreshWork ? currentEpoch + 1 : currentEpoch;
 
         // Use custom async enumerable to avoid exceptions on cancellation.
         NonThrowingChannelReaderAsyncEnumerable<WorkflowEvent> eventStream = new(this._eventChannel.Reader);

dotnet/tests/Microsoft.Agents.AI.Workflows.UnitTests/InProcessExecutionTests.cs

@@ -132,6 +132,53 @@ public class InProcessExecutionTests
             "both versions should produce the same number of agent events");
     }
 
+    /// <summary>
+    /// This test checks that the logic around waiting for input and halting appropriately works right when the
+    /// workflow runs to halting before the EventStream is watched by the user.
+    /// </summary>
+    [Fact]
+    public async Task RunStreamingAsyncWaitToTakeStreamAsync()
+    {
+        // Arrange: Create a simple agent that responds to messages
+        var agent = new SimpleTestAgent("test-agent");
+        var workflow = AgentWorkflowBuilder.BuildSequential(agent);
+        var inputMessage = new ChatMessage(ChatRole.User, "Hello");
+
+        // Act: Execute using streaming version with TurnToken
+        await using StreamingRun run = await InProcessExecution.RunStreamingAsync(workflow, new List<ChatMessage> { inputMessage });
+
+        // Send TurnToken to actually trigger execution (this is the key step)
+        bool messageSent = await run.TrySendMessageAsync(new TurnToken(emitEvents: true));
+        messageSent.Should().BeTrue("TurnToken should be accepted");
+
+        while (await run.GetStatusAsync() != RunStatus.Idle)
+        {
+            await Task.Delay(200);
+        }
+
+        // Collect events
+        List<WorkflowEvent> events = [];
+
+        await foreach (WorkflowEvent evt in run.WatchStreamAsync())
+        {
+            events.Add(evt);
+        }
+
+        // Assert: The workflow should have executed and produced events
+        RunStatus status = await run.GetStatusAsync();
+        status.Should().Be(RunStatus.Idle, "workflow should complete execution");
+
+        events.Should().NotBeEmpty("workflow should produce events during execution");
+
+        // Check that we have agent execution events
+        var agentEvents = events.OfType<AgentResponseUpdateEvent>().ToList();
+        agentEvents.Should().NotBeEmpty("agent should have executed and produced update events");
+
+        // Check that we have output events
+        var outputEvents = events.OfType<WorkflowOutputEvent>().ToList();
+        outputEvents.Should().NotBeEmpty("workflow should produce output events");
+    }
+
     /// <summary>
     /// Simple test agent that echoes back the input message.
     /// </summary>

python/packages/anthropic/agent_framework_anthropic/_chat_client.py

@@ -1271,8 +1271,8 @@ class RawAnthropicClient(
                         )
                     )
                 case "input_json_delta":
-                    # Skip argument deltas for MCP and server tools — execution is handled server-side.
-                    if self._last_call_content_type in ("mcp_tool_use", "server_tool_use"):
+                    # Skip argument deltas for MCP tools — execution is handled server-side.
+                    if self._last_call_content_type == "mcp_tool_use":
                         pass
                     else:
                         call_id = self._last_call_id_name[0] if self._last_call_id_name else ""

python/packages/anthropic/tests/test_anthropic_client.py

@@ -1123,53 +1123,6 @@ def test_parse_contents_from_anthropic_input_json_delta_no_duplicate_name(
     assert result[0].arguments == '"San Francisco"}'
 
 
-def test_parse_contents_server_tool_use_input_json_delta_ignored(
-    mock_anthropic_client: MagicMock,
-) -> None:
-    """Regression test: input_json_delta events are ignored after a server_tool_use block.
-
-    Server-managed tools have their execution handled server-side, so streaming
-    input_json_delta events must not produce Content.from_function_call(name='')
-    entries that would cause Anthropic API 400 errors on subsequent turns.
-    """
-    client = create_test_anthropic_client(mock_anthropic_client)
-
-    # Simulate a server_tool_use event that sets _last_call_content_type
-    server_tool_content = MagicMock()
-    server_tool_content.type = "server_tool_use"
-    server_tool_content.id = "srvtool_abc"
-    server_tool_content.name = "web_search"
-    server_tool_content.input = {}
-
-    result = client._parse_contents_from_anthropic([server_tool_content])
-    # server_tool_use falls through to function_call (not mcp_tool_use / code_execution)
-    assert len(result) == 1
-    assert result[0].type == "function_call"
-    assert client._last_call_content_type == "server_tool_use"  # type: ignore[attr-defined]
-
-    # input_json_delta events after server_tool_use must be silently ignored
-    delta_content = MagicMock()
-    delta_content.type = "input_json_delta"
-    delta_content.partial_json = '{"query": "latest news"}'
-
-    result = client._parse_contents_from_anthropic([delta_content])
-    assert result == [], (
-        "input_json_delta after server_tool_use should produce no content, "
-        "but got: %r" % result
-    )
-
-    # A second delta must also be ignored
-    delta_content_2 = MagicMock()
-    delta_content_2.type = "input_json_delta"
-    delta_content_2.partial_json = '{"extra": true}'
-
-    result = client._parse_contents_from_anthropic([delta_content_2])
-    assert result == [], (
-        "subsequent input_json_delta after server_tool_use should also be ignored, "
-        "but got: %r" % result
-    )
-
-
 # Stream Processing Tests
 
 

python/samples/02-agents/observability/advanced_manual_setup_console_output.py

@@ -8,12 +8,11 @@ from typing import Annotated
 from agent_framework import Message, tool
 from agent_framework.foundry import FoundryChatClient
 from agent_framework.observability import enable_instrumentation
-from azure.identity import AzureCliCredential
 from dotenv import load_dotenv
 from opentelemetry._logs import set_logger_provider
 from opentelemetry.metrics import set_meter_provider
 from opentelemetry.sdk._logs import LoggerProvider, LoggingHandler
-from opentelemetry.sdk._logs.export import BatchLogRecordProcessor, ConsoleLogRecordExporter
+from opentelemetry.sdk._logs.export import BatchLogRecordProcessor, ConsoleLogExporter
 from opentelemetry.sdk.metrics import MeterProvider
 from opentelemetry.sdk.metrics.export import ConsoleMetricExporter, PeriodicExportingMetricReader
 from opentelemetry.sdk.resources import Resource
@@ -38,7 +37,7 @@ def setup_logging():
     # Create and set a global logger provider for the application.
     logger_provider = LoggerProvider(resource=resource)
     # Log processors are initialized with an exporter which is responsible
-    logger_provider.add_log_record_processor(BatchLogRecordProcessor(ConsoleLogRecordExporter()))
+    logger_provider.add_log_record_processor(BatchLogRecordProcessor(ConsoleLogExporter()))
     # Sets the global default logger provider
     set_logger_provider(logger_provider)
     # Create a logging handler to write logging records, in OTLP format, to the exporter.
@@ -116,15 +115,11 @@ async def run_chat_client() -> None:
         2 spans with gen_ai.operation.name=execute_tool
 
     """
-    client = FoundryChatClient(credential=AzureCliCredential())
+    client = FoundryChatClient()
     message = "What's the weather in Amsterdam and in Paris?"
     print(f"User: {message}")
     print("Assistant: ", end="")
-    async for chunk in client.get_response(
-        [Message(role="user", text=message)],
-        stream=True,
-        options={"tools": [get_weather]},
-    ):
+    async for chunk in client.get_response([Message(role="user", text=message)], tools=get_weather, stream=True):
         if chunk.text:
             print(chunk.text, end="")
     print("")

python/samples/02-agents/observability/advanced_zero_code.py

@@ -7,7 +7,6 @@ from typing import TYPE_CHECKING, Annotated
 from agent_framework import Message, tool
 from agent_framework.foundry import FoundryChatClient
 from agent_framework.observability import get_tracer
-from azure.identity import AzureCliCredential
 from dotenv import load_dotenv
 from opentelemetry.trace import SpanKind
 from opentelemetry.trace.span import format_trace_id
@@ -91,19 +90,12 @@ async def run_chat_client(client: "SupportsChatGetResponse", stream: bool = Fals
     print(f"User: {message}")
     if stream:
         print("Assistant: ", end="")
-        async for chunk in client.get_response(
-            [Message(role="user", text=message)],
-            stream=True,
-            options={"tools": [get_weather]},
-        ):
+        async for chunk in client.get_response([Message(role="user", text=message)], tools=get_weather, stream=True):
             if chunk.text:
                 print(chunk.text, end="")
         print("")
     else:
-        response = await client.get_response(
-            [Message(role="user", text=message)],
-            options={"tools": [get_weather]},
-        )
+        response = await client.get_response([Message(role="user", text=message)], tools=get_weather)
         print(f"Assistant: {response}")
 
 
@@ -111,7 +103,7 @@ async def main() -> None:
     with get_tracer().start_as_current_span("Zero Code", kind=SpanKind.CLIENT) as current_span:
         print(f"Trace ID: {format_trace_id(current_span.get_span_context().trace_id)}")
 
-        client = FoundryChatClient(credential=AzureCliCredential())
+        client = FoundryChatClient()
 
         await run_chat_client(client, stream=True)
         await run_chat_client(client, stream=False)

python/samples/02-agents/observability/agent_observability.py

@@ -7,7 +7,6 @@ from typing import Annotated
 from agent_framework import Agent, tool
 from agent_framework.foundry import FoundryChatClient
 from agent_framework.observability import configure_otel_providers, get_tracer
-from azure.identity import AzureCliCredential
 from dotenv import load_dotenv
 from opentelemetry.trace import SpanKind
 from opentelemetry.trace.span import format_trace_id
@@ -19,12 +18,6 @@ load_dotenv()
 """
 This sample shows how you can observe an agent in Agent Framework by using the
 same observability setup function.
-
-Pre-requisites:
-- A Foundry project
-- An observability backend to receive traces and metrics (for example, a local or remote
-  OpenTelemetry Collector, another OTLP-compatible backend, or console exporters enabled
-  via environment variables).
 """
 
 
@@ -54,7 +47,7 @@ async def main():
         print(f"Trace ID: {format_trace_id(current_span.get_span_context().trace_id)}")
 
         agent = Agent(
-            client=FoundryChatClient(credential=AzureCliCredential()),
+            client=FoundryChatClient(),
             tools=get_weather,
             name="WeatherAgent",
             instructions="You are a weather assistant.",

python/samples/02-agents/observability/configure_otel_providers_with_env_var.py

@@ -9,7 +9,6 @@ from typing import TYPE_CHECKING, Annotated, Literal
 from agent_framework import Message, tool
 from agent_framework.foundry import FoundryChatClient
 from agent_framework.observability import configure_otel_providers, get_tracer
-from azure.identity import AzureCliCredential
 from dotenv import load_dotenv
 from opentelemetry import trace
 from opentelemetry.trace.span import format_trace_id
@@ -25,9 +24,8 @@ This sample shows how you can configure observability of an application via the
 When you run this sample with an OTLP endpoint or an Application Insights connection string,
 you should see traces, logs, and metrics in the configured backend.
 
-Pre-requisites:
-- A Foundry project
-- A local OpenTelemetry Collector instance to receive the traces and metrics.
+If no OTLP endpoint or Application Insights connection string is configured, the sample will
+output traces, logs, and metrics to the console.
 """
 
 # Load environment variables from .env file
@@ -80,18 +78,13 @@ async def run_chat_client(client: "SupportsChatGetResponse", stream: bool = Fals
         if stream:
             print("Assistant: ", end="")
             async for chunk in client.get_response(
-                [Message(role="user", text=message)],
-                stream=True,
-                options={"tools": [get_weather]},
+                [Message(role="user", text=message)], tools=get_weather, stream=True
             ):
                 if chunk.text:
                     print(chunk.text, end="")
             print("")
         else:
-            response = await client.get_response(
-                [Message(role="user", text=message)],
-                options={"tools": [get_weather]},
-            )
+            response = await client.get_response([Message(role="user", text=message)], tools=get_weather)
             print(f"Assistant: {response}")
 
 
@@ -108,7 +101,7 @@ async def run_tool() -> None:
     with get_tracer().start_as_current_span("Scenario: AI Function", kind=trace.SpanKind.CLIENT):
         print("Running scenario: AI Function")
         weather = await get_weather.invoke(location="Amsterdam")
-        print(f"Weather in Amsterdam:\n{weather[-1]}")
+        print(f"Weather in Amsterdam:\n{weather}")
 
 
 async def main(scenario: Literal["client", "client_stream", "tool", "all"] = "all"):
@@ -121,7 +114,7 @@ async def main(scenario: Literal["client", "client_stream", "tool", "all"] = "al
     with get_tracer().start_as_current_span("Sample Scenarios", kind=trace.SpanKind.CLIENT) as current_span:
         print(f"Trace ID: {format_trace_id(current_span.get_span_context().trace_id)}")
 
-        client = FoundryChatClient(credential=AzureCliCredential())
+        client = FoundryChatClient()
 
         # Scenarios where telemetry is collected in the SDK, from the most basic to the most complex.
         if scenario == "tool" or scenario == "all":

python/samples/02-agents/observability/configure_otel_providers_with_parameters.py

@@ -10,7 +10,6 @@ from typing import TYPE_CHECKING, Annotated, Literal
 from agent_framework import Message, tool
 from agent_framework.foundry import FoundryChatClient
 from agent_framework.observability import configure_otel_providers, get_tracer
-from azure.identity import AzureCliCredential
 from dotenv import load_dotenv
 from opentelemetry import trace
 from opentelemetry.trace.span import format_trace_id
@@ -28,10 +27,6 @@ and allows you to add multiple exporters programmatically.
 
 For standard OTLP setup, it's recommended to use environment variables (see configure_otel_providers_with_env_var.py).
 Use this approach when you need custom exporter configuration beyond what environment variables provide.
-
-Pre-requisites:
-- A Foundry project
-- A local OpenTelemetry Collector instance to receive the traces and metrics.
 """
 
 # Load environment variables from .env file
@@ -84,18 +79,13 @@ async def run_chat_client(client: "SupportsChatGetResponse", stream: bool = Fals
         if stream:
             print("Assistant: ", end="")
             async for chunk in client.get_response(
-                [Message(role="user", text=message)],
-                stream=True,
-                options={"tools": [get_weather]},
+                [Message(role="user", text=message)], stream=True, tools=get_weather
             ):
                 if chunk.text:
                     print(chunk.text, end="")
             print("")
         else:
-            response = await client.get_response(
-                [Message(role="user", text=message)],
-                options={"tools": [get_weather]},
-            )
+            response = await client.get_response([Message(role="user", text=message)], tools=get_weather)
             print(f"Assistant: {response}")
 
 
@@ -112,7 +102,7 @@ async def run_tool() -> None:
     with get_tracer().start_as_current_span("Scenario: AI Function", kind=trace.SpanKind.CLIENT):
         print("Running scenario: AI Function")
         weather = await get_weather.invoke(location="Amsterdam")
-        print(f"Weather in Amsterdam:\n{weather[-1]}")
+        print(f"Weather in Amsterdam:\n{weather}")
 
 
 async def main(scenario: Literal["client", "client_stream", "tool", "all"] = "all"):
@@ -163,7 +153,7 @@ async def main(scenario: Literal["client", "client_stream", "tool", "all"] = "al
     with get_tracer().start_as_current_span("Sample Scenarios", kind=trace.SpanKind.CLIENT) as current_span:
         print(f"Trace ID: {format_trace_id(current_span.get_span_context().trace_id)}")
 
-        client = FoundryChatClient(credential=AzureCliCredential())
+        client = FoundryChatClient()
 
         # Scenarios where telemetry is collected in the SDK, from the most basic to the most complex.
         if scenario == "tool" or scenario == "all":

python/samples/02-agents/providers/anthropic/anthropic_claude_with_mcp.py

@@ -12,9 +12,6 @@ Supported MCP server types:
 - "http": Remote HTTP server
 - "sse": Remote SSE (Server-Sent Events) server
 
-Environment variables:
-- ANTHROPIC_API_KEY: Your Anthropic API key
-
 SECURITY NOTE: MCP servers can expose powerful capabilities. Only configure
 servers you trust. Use permission handlers to control what actions are allowed.
 """

python/samples/02-agents/providers/anthropic/anthropic_claude_with_multiple_permissions.py

@@ -15,9 +15,6 @@ Available built-in tools:
 - "Glob": Search for files by pattern
 - "Grep": Search file contents
 
-Environment variables:
-- ANTHROPIC_API_KEY: Your Anthropic API key
-
 SECURITY NOTE: Only enable permissions that are necessary for your use case.
 More permissions mean more potential for unintended actions.
 """
@@ -27,10 +24,6 @@ from typing import Any
 
 from agent_framework.anthropic import ClaudeAgent
 from claude_agent_sdk import PermissionResultAllow, PermissionResultDeny
-from dotenv import load_dotenv
-
-# Load environment variables from .env file
-load_dotenv()
 
 
 async def prompt_permission(

python/samples/02-agents/providers/anthropic/anthropic_claude_with_session.py

@@ -6,9 +6,6 @@ Claude Agent with Session Management
 This sample demonstrates session management with ClaudeAgent, showing
 persistent conversation capabilities. Sessions are automatically persisted
 by the Claude Code CLI.
-
-Environment variables:
-- ANTHROPIC_API_KEY: Your Anthropic API key
 """
 
 import asyncio
@@ -17,12 +14,8 @@ from typing import Annotated
 
 from agent_framework import tool
 from agent_framework.anthropic import ClaudeAgent
-from dotenv import load_dotenv
 from pydantic import Field
 
-# Load environment variables from .env file
-load_dotenv()
-
 
 @tool
 def get_weather(

python/samples/02-agents/providers/anthropic/anthropic_claude_with_shell.py

@@ -7,9 +7,6 @@ This sample demonstrates how to enable shell command execution with ClaudeAgent.
 By providing a permission handler via `can_use_tool`, the agent can execute
 shell commands to perform tasks like listing files, running scripts, or executing system commands.
 
-Environment variables:
-- ANTHROPIC_API_KEY: Your Anthropic API key
-
 SECURITY NOTE: Only enable shell permissions when you trust the agent's actions.
 Shell commands have full access to your system within the permissions of the running process.
 """
@@ -19,10 +16,6 @@ from typing import Any
 
 from agent_framework.anthropic import ClaudeAgent
 from claude_agent_sdk import PermissionResultAllow, PermissionResultDeny
-from dotenv import load_dotenv
-
-# Load environment variables from .env file
-load_dotenv()
 
 
 async def prompt_permission(

python/samples/02-agents/providers/anthropic/anthropic_claude_with_tools.py

@@ -13,18 +13,11 @@ Available built-in tools:
 - "Edit": Edit existing files
 - "Glob": Search for files by pattern
 - "Grep": Search file contents
-
-Environment variables:
-- ANTHROPIC_API_KEY: Your Anthropic API key
 """
 
 import asyncio
 
 from agent_framework.anthropic import ClaudeAgent
-from dotenv import load_dotenv
-
-# Load environment variables from .env file
-load_dotenv()
 
 
 async def main() -> None:

python/samples/02-agents/providers/anthropic/anthropic_claude_with_url.py

@@ -10,9 +10,6 @@ Available web tools:
 - "WebFetch": Fetch content from URLs
 - "WebSearch": Search the web
 
-Environment variables:
-- ANTHROPIC_API_KEY: Your Anthropic API key
-
 SECURITY NOTE: Only enable URL permissions when you trust the agent's actions.
 URL fetching allows the agent to access any URL accessible from your network.
 """
@@ -20,10 +17,6 @@ URL fetching allows the agent to access any URL accessible from your network.
 import asyncio
 
 from agent_framework.anthropic import ClaudeAgent
-from dotenv import load_dotenv
-
-# Load environment variables from .env file
-load_dotenv()
 
 
 async def main() -> None:

python/samples/02-agents/providers/anthropic/anthropic_skills.py

@@ -21,10 +21,6 @@ This sample demonstrates using Anthropic with:
     You can also set additonal_chat_options with "additional_beta_flags" per request.
 - Creating an agent with the Code Interpreter tool and a Skill.
 - Catching and downloading generated files from the agent.
-
-Environment variables:
-- ANTHROPIC_API_KEY: Your Anthropic API key
-- ANTHROPIC_CHAT_MODEL_ID: The Anthropic model to use, such as "claude-sonnet-4-5-20250929"
 """