From e6344888335da039bd66b24e9d35f1e548f281fc Mon Sep 17 00:00:00 2001
From: Tao Chen <taochen@microsoft.com>
Date: Tue, 7 Apr 2026 20:44:30 -0700
Subject: [PATCH 01/13] Experiment

---
 python/.cspell.json                           |   1 +
 .../core/agent_framework/foundry/__init__.py  |   2 +
 .../core/agent_framework/foundry/__init__.pyi |   2 +
 python/packages/core/pyproject.toml           |   1 +
 python/packages/foundry_hosting/LICENSE       |  21 ++
 python/packages/foundry_hosting/README.md     |   0
 .../__init__.py                               |  12 ++
 .../_responses.py                             | 109 +++++++++++
 .../_shared.py                                | 181 ++++++++++++++++++
 .../packages/foundry_hosting/pyproject.toml   |  99 ++++++++++
 python/pyproject.toml                         |   9 +
 .../foundry_vnext/helloworld/.dockerignore    |   6 +
 .../foundry_vnext/helloworld/Dockerfile       |  16 ++
 .../helloworld/agent.manifest.yaml            |  16 ++
 .../foundry_vnext/helloworld/agent.yaml       |   8 +
 .../foundry_vnext/helloworld/main.py          |  33 ++++
 .../foundry_vnext/helloworld/requirements.txt |   2 +
 python/uv.lock                                | 123 +++++++++++-
 18 files changed, 637 insertions(+), 4 deletions(-)
 create mode 100644 python/packages/foundry_hosting/LICENSE
 create mode 100644 python/packages/foundry_hosting/README.md
 create mode 100644 python/packages/foundry_hosting/agent_framework_foundry_hosting/__init__.py
 create mode 100644 python/packages/foundry_hosting/agent_framework_foundry_hosting/_responses.py
 create mode 100644 python/packages/foundry_hosting/agent_framework_foundry_hosting/_shared.py
 create mode 100644 python/packages/foundry_hosting/pyproject.toml
 create mode 100644 python/samples/05-end-to-end/hosted_agents/foundry_vnext/helloworld/.dockerignore
 create mode 100644 python/samples/05-end-to-end/hosted_agents/foundry_vnext/helloworld/Dockerfile
 create mode 100644 python/samples/05-end-to-end/hosted_agents/foundry_vnext/helloworld/agent.manifest.yaml
 create mode 100644 python/samples/05-end-to-end/hosted_agents/foundry_vnext/helloworld/agent.yaml
 create mode 100644 python/samples/05-end-to-end/hosted_agents/foundry_vnext/helloworld/main.py
 create mode 100644 python/samples/05-end-to-end/hosted_agents/foundry_vnext/helloworld/requirements.txt

diff --git a/python/.cspell.json b/python/.cspell.json
index a26cc7fed7..c5bf954d2b 100644
--- a/python/.cspell.json
+++ b/python/.cspell.json
@@ -24,6 +24,7 @@
     ],
     "words": [
         "aeiou",
+        "agentserver",
         "agui",
         "aiplatform",
         "azuredocindex",
diff --git a/python/packages/core/agent_framework/foundry/__init__.py b/python/packages/core/agent_framework/foundry/__init__.py
index b1d2b88450..8edaed5485 100644
--- a/python/packages/core/agent_framework/foundry/__init__.py
+++ b/python/packages/core/agent_framework/foundry/__init__.py
@@ -6,6 +6,7 @@
 - ``agent-framework-anthropic``
 - ``agent-framework-foundry``
 - ``agent-framework-foundry-local``
+- ``agent-framework-foundry-hosting``
 """
 
 import importlib
@@ -31,6 +32,7 @@
     "RawFoundryEmbeddingClient": ("agent_framework_foundry", "agent-framework-foundry"),
     "evaluate_foundry_target": ("agent_framework_foundry", "agent-framework-foundry"),
     "evaluate_traces": ("agent_framework_foundry", "agent-framework-foundry"),
+    "ResponsesHost": ("agent_framework_foundry_hosting", "agent-framework-foundry-hosting"),
 }
 
 
diff --git a/python/packages/core/agent_framework/foundry/__init__.pyi b/python/packages/core/agent_framework/foundry/__init__.pyi
index 47eb92b3af..133378ea2a 100644
--- a/python/packages/core/agent_framework/foundry/__init__.pyi
+++ b/python/packages/core/agent_framework/foundry/__init__.pyi
@@ -20,6 +20,7 @@ from agent_framework_foundry import (
     evaluate_foundry_target,
     evaluate_traces,
 )
+from agent_framework_foundry_hosting import ResponsesHost
 from agent_framework_foundry_local import (
     FoundryLocalChatOptions,
     FoundryLocalClient,
@@ -44,6 +45,7 @@ __all__ = [
     "RawFoundryAgentChatClient",
     "RawFoundryChatClient",
     "RawFoundryEmbeddingClient",
+    "ResponsesHost",
     "evaluate_foundry_target",
     "evaluate_traces",
 ]
diff --git a/python/packages/core/pyproject.toml b/python/packages/core/pyproject.toml
index e4aa1f5e40..c6e7ef2b42 100644
--- a/python/packages/core/pyproject.toml
+++ b/python/packages/core/pyproject.toml
@@ -48,6 +48,7 @@ all = [
     "agent-framework-durabletask",
     "agent-framework-foundry",
     "agent-framework-foundry-local",
+    "agent-framework-foundry-hosting",
     "agent-framework-github-copilot; python_version >= '3.11'",
     "agent-framework-lab",
     "agent-framework-mem0",
diff --git a/python/packages/foundry_hosting/LICENSE b/python/packages/foundry_hosting/LICENSE
new file mode 100644
index 0000000000..9e841e7a26
--- /dev/null
+++ b/python/packages/foundry_hosting/LICENSE
@@ -0,0 +1,21 @@
+    MIT License
+
+    Copyright (c) Microsoft Corporation.
+
+    Permission is hereby granted, free of charge, to any person obtaining a copy
+    of this software and associated documentation files (the "Software"), to deal
+    in the Software without restriction, including without limitation the rights
+    to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+    copies of the Software, and to permit persons to whom the Software is
+    furnished to do so, subject to the following conditions:
+
+    The above copyright notice and this permission notice shall be included in all
+    copies or substantial portions of the Software.
+
+    THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+    IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+    FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+    AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+    LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+    OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+    SOFTWARE
diff --git a/python/packages/foundry_hosting/README.md b/python/packages/foundry_hosting/README.md
new file mode 100644
index 0000000000..e69de29bb2
diff --git a/python/packages/foundry_hosting/agent_framework_foundry_hosting/__init__.py b/python/packages/foundry_hosting/agent_framework_foundry_hosting/__init__.py
new file mode 100644
index 0000000000..595bdc8e1b
--- /dev/null
+++ b/python/packages/foundry_hosting/agent_framework_foundry_hosting/__init__.py
@@ -0,0 +1,12 @@
+# Copyright (c) Microsoft. All rights reserved.
+
+import importlib.metadata
+
+from ._responses import ResponsesHost
+
+try:
+    __version__ = importlib.metadata.version(__name__)
+except importlib.metadata.PackageNotFoundError:
+    __version__ = "0.0.0"
+
+__all__ = ["ResponsesHost"]
diff --git a/python/packages/foundry_hosting/agent_framework_foundry_hosting/_responses.py b/python/packages/foundry_hosting/agent_framework_foundry_hosting/_responses.py
new file mode 100644
index 0000000000..6e3afff5bc
--- /dev/null
+++ b/python/packages/foundry_hosting/agent_framework_foundry_hosting/_responses.py
@@ -0,0 +1,109 @@
+# Copyright (c) Microsoft. All rights reserved.
+
+import asyncio
+from collections.abc import AsyncIterable
+
+from agent_framework import Agent, HistoryProvider, Message
+from azure.ai.agentserver.core import AgentHost
+from azure.ai.agentserver.responses import ResponseContext, ResponseEventStream
+from azure.ai.agentserver.responses.hosting import ResponseHandler
+from azure.ai.agentserver.responses.models import CreateResponse, get_input_text
+from typing_extensions import Any, Sequence
+
+from ._shared import extract_chat_options, to_messages
+
+
+class ResponsesHostContextProvider(HistoryProvider):
+    def __init__(self, context: ResponseContext):
+        super().__init__("responses-host", load_messages=True)
+        self.context = context
+
+    async def get_messages(
+        self,
+        session_id: str | None,
+        *,
+        state: dict[str, Any] | None = None,
+        **kwargs: Any,
+    ) -> list[Message]:
+        history = await self.context.get_history_async()
+        return to_messages(history)
+
+    async def save_messages(
+        self,
+        session_id: str | None,
+        messages: Sequence[Message],
+        *,
+        state: dict[str, Any] | None = None,
+        **kwargs: Any,
+    ) -> None:
+        pass
+
+
+class ResponsesHost(AgentHost):
+    def __init__(self, agent: Agent, **kwargs: Any) -> None:
+        application_insights_connection_string = kwargs.pop("application_insights_connection_string", None)
+        graceful_shutdown_timeout = kwargs.pop("graceful_shutdown_timeout", None)
+        log_level = kwargs.pop("log_level", None)
+        super().__init__(
+            application_insights_connection_string=application_insights_connection_string,
+            graceful_shutdown_timeout=graceful_shutdown_timeout,
+            log_level=log_level,
+        )
+
+        self.agent = agent
+        self.response_handler = ResponseHandler(self)
+        self.response_handler.create_handler(self._handle_create)  # type: ignore
+
+    async def _handle_create(
+        self,
+        request: CreateResponse,
+        context: ResponseContext,
+        cancellation_signal: asyncio.Event,
+    ) -> AsyncIterable[dict[str, Any]]:
+        # Replace or add a history provider that has `load_messages=True`
+        history_provider_idx: list[int] = []
+        for i, provider in enumerate(self.agent.context_providers):
+            if isinstance(provider, HistoryProvider) and provider.load_messages:
+                history_provider_idx.append(i)
+
+        if not history_provider_idx:
+            self.agent.context_providers.append(ResponsesHostContextProvider(context))
+        elif len(history_provider_idx) > 1:
+            # There shouldn't be more than one history provider with `load_messages=True`
+            raise RuntimeError("There shouldn't be more than one history provider with `load_messages=True`")
+        else:
+            self.agent.context_providers[history_provider_idx[0]] = ResponsesHostContextProvider(context)
+
+        stream = ResponseEventStream(
+            response_id=context.response_id,
+            model=getattr(request, "model", None),
+        )
+
+        yield stream.emit_created()
+        yield stream.emit_in_progress()
+
+        input_items = get_input_text(request)
+
+        # Start the response
+        message_item = stream.add_output_item_message()
+        yield message_item.emit_added()
+        text_content = message_item.add_text_content()
+        yield text_content.emit_added()
+
+        # Invoke the MAF agent
+        chat_options = extract_chat_options(request)
+        full_text = ""
+        async for update in self.agent.run(
+            input_items,
+            options=chat_options,
+            stream=True,
+        ):
+            full_text += update.text
+            yield text_content.emit_delta(update.text)
+
+        # Complete the message
+        yield text_content.emit_done(full_text)
+        yield message_item.emit_content_done(text_content)
+        yield message_item.emit_done()
+
+        yield stream.emit_completed()
diff --git a/python/packages/foundry_hosting/agent_framework_foundry_hosting/_shared.py b/python/packages/foundry_hosting/agent_framework_foundry_hosting/_shared.py
new file mode 100644
index 0000000000..3cd64bed1d
--- /dev/null
+++ b/python/packages/foundry_hosting/agent_framework_foundry_hosting/_shared.py
@@ -0,0 +1,181 @@
+# Copyright (c) Microsoft. All rights reserved.
+
+from agent_framework import ChatOptions, Content, Message
+from azure.ai.agentserver.responses.models import (
+    ComputerScreenshotContent,
+    CreateResponse,
+    MessageContent,
+    MessageContentInputFileContent,
+    MessageContentInputImageContent,
+    MessageContentInputTextContent,
+    MessageContentOutputTextContent,
+    MessageContentReasoningTextContent,
+    MessageContentRefusalContent,
+    OutputItem,
+    OutputItemFunctionToolCall,
+    OutputItemFunctionToolCallOutput,
+    OutputItemMessage,
+    OutputItemOutputMessage,
+    OutputItemReasoningItem,
+    OutputMessageContent,
+    OutputMessageContentOutputTextContent,
+    OutputMessageContentRefusalContent,
+    SummaryTextContent,
+    TextContent,
+)
+from typing_extensions import Sequence, cast
+
+
+def extract_chat_options(request: CreateResponse) -> ChatOptions:
+    """Extracts chat options from a CreateResponse request.
+
+    Args:
+        request (CreateResponse): The CreateResponse request object containing the chat options.
+
+    Returns:
+        ChatOptions: The extracted chat options.
+    """
+    options = ChatOptions()
+
+    if request.temperature is not None:
+        options["temperature"] = request.temperature
+    if request.top_p is not None:
+        options["top_p"] = request.top_p
+    if request.max_output_tokens is not None:
+        options["max_tokens"] = request.max_output_tokens
+
+    return options
+
+
+def to_messages(history: Sequence[OutputItem]) -> list[Message]:
+    """Converts a sequence of OutputItem objects to a list of Message objects.
+
+    Args:
+        history (Sequence[OutputItem]): The sequence of OutputItem objects to convert.
+
+    Returns:
+        list[Message]: The list of Message objects.
+    """
+    messages: list[Message] = []
+    for item in history:
+        messages.append(_to_message(item))
+    return messages
+
+
+def _to_message(item: OutputItem) -> Message:
+    """Converts an OutputItem to a Message.
+
+    Args:
+        item (OutputItem): The OutputItem to convert.
+
+    Returns:
+        Message: The converted Message.
+
+    Raises:
+        ValueError: If the OutputItem type is not supported.
+    """
+    if item.type == "output_message":
+        msg = cast(OutputItemOutputMessage, item)
+        contents = [_convert_output_message_content(part) for part in msg.content]
+        return Message(role=msg.role, contents=contents)
+
+    if item.type == "message":
+        msg = cast(OutputItemMessage, item)
+        contents = [_convert_message_content(part) for part in msg.content]
+        return Message(role=msg.role, contents=contents)
+
+    if item.type == "function_call":
+        fc = cast(OutputItemFunctionToolCall, item)
+        return Message(
+            role="assistant",
+            contents=[Content.from_function_call(fc.call_id, fc.name, arguments=fc.arguments)],
+        )
+
+    if item.type == "function_call_output":
+        fco = cast(OutputItemFunctionToolCallOutput, item)
+        output = fco.output if isinstance(fco.output, str) else str(fco.output)
+        return Message(
+            role="tool",
+            contents=[Content.from_function_result(fco.call_id, result=output)],
+        )
+
+    if item.type == "reasoning":
+        reasoning = cast(OutputItemReasoningItem, item)
+        contents: list[Content] = []
+        if reasoning.summary:
+            for summary in reasoning.summary:
+                contents.append(Content.from_text(summary.text))
+        return Message(role="assistant", contents=contents)
+
+    raise ValueError(f"Unsupported OutputItem type: {item.type}")
+
+
+def _convert_output_message_content(content: OutputMessageContent) -> Content:
+    """Converts an OutputMessageContent to a Content object.
+
+    Args:
+        content (OutputMessageContent): The OutputMessageContent to convert.
+
+    Returns:
+        Content: The converted Content object.
+
+    Raises:
+        ValueError: If the OutputMessageContent type is not supported.
+    """
+    if content.type == "output_text":
+        text_content = cast(OutputMessageContentOutputTextContent, content)
+        return Content.from_text(text_content.text)
+    if content.type == "refusal":
+        refusal_content = cast(OutputMessageContentRefusalContent, content)
+        return Content.from_text(refusal_content.refusal)
+
+    raise ValueError(f"Unsupported OutputMessageContent type: {content.type}")
+
+
+def _convert_message_content(content: MessageContent) -> Content:
+    """Converts a MessageContent to a Content object.
+
+    Args:
+        content (MessageContent): The MessageContent to convert.
+
+    Returns:
+        Content: The converted Content object.
+
+    Raises:
+        ValueError: If the MessageContent type is not supported.
+    """
+    if content.type == "input_text":
+        input_text = cast(MessageContentInputTextContent, content)
+        return Content.from_text(input_text.text)
+    if content.type == "output_text":
+        output_text = cast(MessageContentOutputTextContent, content)
+        return Content.from_text(output_text.text)
+    if content.type == "text":
+        text = cast(TextContent, content)
+        return Content.from_text(text.text)
+    if content.type == "summary_text":
+        summary = cast(SummaryTextContent, content)
+        return Content.from_text(summary.text)
+    if content.type == "refusal":
+        refusal = cast(MessageContentRefusalContent, content)
+        return Content.from_text(refusal.refusal)
+    if content.type == "reasoning_text":
+        reasoning = cast(MessageContentReasoningTextContent, content)
+        return Content.from_text_reasoning(text=reasoning.text)
+    if content.type == "input_image":
+        image = cast(MessageContentInputImageContent, content)
+        if image.image_url:
+            return Content.from_uri(image.image_url)
+        if image.file_id:
+            return Content.from_hosted_file(image.file_id)
+    if content.type == "input_file":
+        file = cast(MessageContentInputFileContent, content)
+        if file.file_url:
+            return Content.from_uri(file.file_url)
+        if file.file_id:
+            return Content.from_hosted_file(file.file_id, name=file.filename)
+    if content.type == "computer_screenshot":
+        screenshot = cast(ComputerScreenshotContent, content)
+        return Content.from_uri(screenshot.image_url)
+
+    raise ValueError(f"Unsupported MessageContent type: {content.type}")
diff --git a/python/packages/foundry_hosting/pyproject.toml b/python/packages/foundry_hosting/pyproject.toml
new file mode 100644
index 0000000000..bce8bbfd69
--- /dev/null
+++ b/python/packages/foundry_hosting/pyproject.toml
@@ -0,0 +1,99 @@
+[project]
+name = "agent-framework-foundry-hosting"
+description = "Foundry Local integration for Microsoft Agent Framework."
+authors = [{ name = "Microsoft", email = "af-support@microsoft.com"}]
+readme = "README.md"
+requires-python = ">=3.10"
+version = "1.0.0b260402"
+license-files = ["LICENSE"]
+urls.homepage = "https://aka.ms/agent-framework"
+urls.source = "https://github.com/microsoft/agent-framework/tree/main/python"
+urls.release_notes = "https://github.com/microsoft/agent-framework/releases?q=tag%3Apython-1&expanded=true"
+urls.issues = "https://github.com/microsoft/agent-framework/issues"
+classifiers = [
+  "License :: OSI Approved :: MIT License",
+  "Development Status :: 4 - Beta",
+  "Intended Audience :: Developers",
+  "Programming Language :: Python :: 3",
+  "Programming Language :: Python :: 3.10",
+  "Programming Language :: Python :: 3.11",
+  "Programming Language :: Python :: 3.12",
+  "Programming Language :: Python :: 3.13",
+  "Programming Language :: Python :: 3.14",
+  "Typing :: Typed",
+]
+dependencies = [
+    "agent-framework-core>=1.0.0,<2",
+    "azure-ai-agentserver-core==2.0.0a20260331006",
+    "azure-ai-agentserver-responses==1.0.0a20260331006",
+    "azure-ai-agentserver-invocations==1.0.0a20260331006"
+]
+
+[tool.uv]
+prerelease = "if-necessary-or-explicit"
+environments = [
+    "sys_platform == 'darwin'",
+    "sys_platform == 'linux'",
+    "sys_platform == 'win32'"
+]
+
+[tool.uv-dynamic-versioning]
+fallback-version = "0.0.0"
+
+[tool.pytest.ini_options]
+testpaths = 'tests'
+addopts = "-ra -q -r fEX"
+asyncio_mode = "auto"
+asyncio_default_fixture_loop_scope = "function"
+filterwarnings = []
+timeout = 120
+markers = [
+    "integration: marks tests as integration tests that require external services",
+]
+
+[tool.ruff]
+extend = "../../pyproject.toml"
+
+[tool.coverage.run]
+omit = [
+    "**/__init__.py"
+]
+
+[tool.pyright]
+extends = "../../pyproject.toml"
+include = ["agent_framework_foundry_hosting"]
+exclude = ['tests']
+
+[tool.mypy]
+plugins = ['pydantic.mypy']
+strict = true
+python_version = "3.10"
+ignore_missing_imports = true
+disallow_untyped_defs = true
+no_implicit_optional = true
+check_untyped_defs = true
+warn_return_any = true
+show_error_codes = true
+warn_unused_ignores = false
+disallow_incomplete_defs = true
+disallow_untyped_decorators = true
+
+[tool.bandit]
+targets = ["agent_framework_foundry_hosting"]
+exclude_dirs = ["tests"]
+
+[tool.poe]
+executor.type = "uv"
+include = "../../shared_tasks.toml"
+
+[tool.poe.tasks.mypy]
+help = "Run MyPy for this package."
+cmd = "mypy --config-file $POE_ROOT/pyproject.toml agent_framework_foundry_hosting"
+
+[tool.poe.tasks.test]
+help = "Run the default unit test suite for this package."
+cmd = 'pytest -m "not integration" --cov=agent_framework_foundry_hosting --cov-report=term-missing:skip-covered tests'
+
+[build-system]
+requires = ["flit-core >= 3.11,<4.0"]
+build-backend = "flit_core.buildapi"
\ No newline at end of file
diff --git a/python/pyproject.toml b/python/pyproject.toml
index 24af13b940..c2b692e23c 100644
--- a/python/pyproject.toml
+++ b/python/pyproject.toml
@@ -78,6 +78,7 @@ agent-framework-declarative = { workspace = true }
 agent-framework-devui = { workspace = true }
 agent-framework-durabletask = { workspace = true }
 agent-framework-foundry = { workspace = true }
+agent-framework-foundry-hosting = { workspace = true }
 agent-framework-foundry-local = { workspace = true }
 agent-framework-lab = { workspace = true }
 agent-framework-mem0 = { workspace = true }
@@ -88,6 +89,9 @@ agent-framework-redis = { workspace = true }
 agent-framework-github-copilot = { workspace = true }
 agent-framework-claude = { workspace = true }
 agent-framework-orchestrations = { workspace = true }
+azure-ai-agentserver-responses = { index = "azure-sdk-feed" }
+azure-ai-agentserver-invocations = { index = "azure-sdk-feed" }
+azure-ai-agentserver-core = { index = "azure-sdk-feed" }
 litellm = { url = "https://files.pythonhosted.org/packages/57/77/0c6eca2cb049793ddf8ce9cdcd5123a35666c4962514788c4fc90edf1d3b/litellm-1.82.1-py3-none-any.whl" }
 
 [tool.ruff]
@@ -418,6 +422,11 @@ url = "https://test.pypi.org/simple/"
 publish-url = "https://test.pypi.org/legacy/"
 explicit = true
 
+[[tool.uv.index]]
+name = "azure-sdk-feed"
+url = "https://pkgs.dev.azure.com/azure-sdk/public/_packaging/azure-sdk-for-python/pypi/simple/"
+explicit = true
+
 [tool.flit.module]
 name = "agent_framework_meta"
 
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext/helloworld/.dockerignore b/python/samples/05-end-to-end/hosted_agents/foundry_vnext/helloworld/.dockerignore
new file mode 100644
index 0000000000..008e6e6616
--- /dev/null
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_vnext/helloworld/.dockerignore
@@ -0,0 +1,6 @@
+.venv
+__pycache__
+*.pyc
+*.pyo
+*.pyd
+.Python
\ No newline at end of file
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext/helloworld/Dockerfile b/python/samples/05-end-to-end/hosted_agents/foundry_vnext/helloworld/Dockerfile
new file mode 100644
index 0000000000..eaffb94f19
--- /dev/null
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_vnext/helloworld/Dockerfile
@@ -0,0 +1,16 @@
+FROM python:3.12-slim
+
+WORKDIR /app
+
+COPY . user_agent/
+WORKDIR /app/user_agent
+
+RUN if [ -f requirements.txt ]; then \
+        pip install -r requirements.txt; \
+    else \
+        echo "No requirements.txt found"; \
+    fi
+
+EXPOSE 8088
+
+CMD ["python", "main.py"]
\ No newline at end of file
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext/helloworld/agent.manifest.yaml b/python/samples/05-end-to-end/hosted_agents/foundry_vnext/helloworld/agent.manifest.yaml
new file mode 100644
index 0000000000..ad074d6b89
--- /dev/null
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_vnext/helloworld/agent.manifest.yaml
@@ -0,0 +1,16 @@
+name: echo-agent-responses-streaming
+description: >
+  A simple echo agent that streams responses word-by-word using the
+  azure-ai-agentserver-responses SDK with Server-Sent Events (SSE).
+metadata:
+  tags:
+    - AI Agent Hosting
+    - Azure AI AgentServer
+    - Responses Protocol
+    - Streaming
+template:
+  name: echo-agent-responses-streaming
+  kind: hosted
+  protocols:
+    - protocol: responses
+      version: v0.1.0
\ No newline at end of file
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext/helloworld/agent.yaml b/python/samples/05-end-to-end/hosted_agents/foundry_vnext/helloworld/agent.yaml
new file mode 100644
index 0000000000..3738c690a1
--- /dev/null
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_vnext/helloworld/agent.yaml
@@ -0,0 +1,8 @@
+kind: hosted
+name: echo-agent-responses-streaming
+protocols:
+  - protocol: responses
+    version: v0.1.0
+resources:
+  cpu: "0.25"
+  memory: 0.5Gi
\ No newline at end of file
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext/helloworld/main.py b/python/samples/05-end-to-end/hosted_agents/foundry_vnext/helloworld/main.py
new file mode 100644
index 0000000000..f1e1972393
--- /dev/null
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_vnext/helloworld/main.py
@@ -0,0 +1,33 @@
+# Copyright (c) Microsoft. All rights reserved.
+
+
+import os
+
+from agent_framework import Agent
+from agent_framework.foundry import FoundryChatClient, ResponsesHost
+from azure.identity import AzureCliCredential
+from dotenv import load_dotenv
+
+# Load environment variables from .env file
+load_dotenv()
+
+
+def main():
+    client = FoundryChatClient(
+        project_endpoint=os.environ["FOUNDRY_PROJECT_ENDPOINT"],
+        model=os.environ["FOUNDRY_MODEL"],
+        credential=AzureCliCredential(),
+    )
+
+    agent = Agent(
+        client=client,
+        name="HelloAgent",
+        instructions="You are a friendly assistant. Keep your answers brief.",
+    )
+
+    server = ResponsesHost(agent)
+    server.run()
+
+
+if __name__ == "__main__":
+    main()
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext/helloworld/requirements.txt b/python/samples/05-end-to-end/hosted_agents/foundry_vnext/helloworld/requirements.txt
new file mode 100644
index 0000000000..f7dc62f3e3
--- /dev/null
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_vnext/helloworld/requirements.txt
@@ -0,0 +1,2 @@
+agent-framework
+agent-framework-foundry-hosting
\ No newline at end of file
diff --git a/python/uv.lock b/python/uv.lock
index fef2474e59..a713525a20 100644
--- a/python/uv.lock
+++ b/python/uv.lock
@@ -42,6 +42,7 @@ members = [
     "agent-framework-devui",
     "agent-framework-durabletask",
     "agent-framework-foundry",
+    "agent-framework-foundry-hosting",
     "agent-framework-foundry-local",
     "agent-framework-github-copilot",
     "agent-framework-lab",
@@ -343,6 +344,7 @@ all = [
     { name = "agent-framework-devui", marker = "sys_platform == 'darwin' or sys_platform == 'linux' or sys_platform == 'win32'" },
     { name = "agent-framework-durabletask", marker = "sys_platform == 'darwin' or sys_platform == 'linux' or sys_platform == 'win32'" },
     { name = "agent-framework-foundry", marker = "sys_platform == 'darwin' or sys_platform == 'linux' or sys_platform == 'win32'" },
+    { name = "agent-framework-foundry-hosting", marker = "sys_platform == 'darwin' or sys_platform == 'linux' or sys_platform == 'win32'" },
     { name = "agent-framework-foundry-local", marker = "sys_platform == 'darwin' or sys_platform == 'linux' or sys_platform == 'win32'" },
     { name = "agent-framework-github-copilot", marker = "(python_full_version >= '3.11' and sys_platform == 'darwin') or (python_full_version >= '3.11' and sys_platform == 'linux') or (python_full_version >= '3.11' and sys_platform == 'win32')" },
     { name = "agent-framework-lab", marker = "sys_platform == 'darwin' or sys_platform == 'linux' or sys_platform == 'win32'" },
@@ -371,6 +373,7 @@ requires-dist = [
     { name = "agent-framework-devui", marker = "extra == 'all'", editable = "packages/devui" },
     { name = "agent-framework-durabletask", marker = "extra == 'all'", editable = "packages/durabletask" },
     { name = "agent-framework-foundry", marker = "extra == 'all'", editable = "packages/foundry" },
+    { name = "agent-framework-foundry-hosting", marker = "extra == 'all'", editable = "packages/foundry_hosting" },
     { name = "agent-framework-foundry-local", marker = "extra == 'all'", editable = "packages/foundry_local" },
     { name = "agent-framework-github-copilot", marker = "python_full_version >= '3.11' and extra == 'all'", editable = "packages/github_copilot" },
     { name = "agent-framework-lab", marker = "extra == 'all'", editable = "packages/lab" },
@@ -497,6 +500,25 @@ requires-dist = [
     { name = "azure-ai-projects", specifier = ">=2.0.0,<3.0" },
 ]
 
+[[package]]
+name = "agent-framework-foundry-hosting"
+version = "1.0.0b260402"
+source = { editable = "packages/foundry_hosting" }
+dependencies = [
+    { name = "agent-framework-core", marker = "sys_platform == 'darwin' or sys_platform == 'linux' or sys_platform == 'win32'" },
+    { name = "azure-ai-agentserver-core", marker = "sys_platform == 'darwin' or sys_platform == 'linux' or sys_platform == 'win32'" },
+    { name = "azure-ai-agentserver-invocations", marker = "sys_platform == 'darwin' or sys_platform == 'linux' or sys_platform == 'win32'" },
+    { name = "azure-ai-agentserver-responses", marker = "sys_platform == 'darwin' or sys_platform == 'linux' or sys_platform == 'win32'" },
+]
+
+[package.metadata]
+requires-dist = [
+    { name = "agent-framework-core", editable = "packages/core" },
+    { name = "azure-ai-agentserver-core", specifier = "==2.0.0a20260331006", index = "https://pkgs.dev.azure.com/azure-sdk/public/_packaging/azure-sdk-for-python/pypi/simple/" },
+    { name = "azure-ai-agentserver-invocations", specifier = "==1.0.0a20260331006", index = "https://pkgs.dev.azure.com/azure-sdk/public/_packaging/azure-sdk-for-python/pypi/simple/" },
+    { name = "azure-ai-agentserver-responses", specifier = "==1.0.0a20260331006", index = "https://pkgs.dev.azure.com/azure-sdk/public/_packaging/azure-sdk-for-python/pypi/simple/" },
+]
+
 [[package]]
 name = "agent-framework-foundry-local"
 version = "1.0.0b260402"
@@ -996,6 +1018,46 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/64/b4/17d4b0b2a2dc85a6df63d1157e028ed19f90d4cd97c36717afef2bc2f395/attrs-26.1.0-py3-none-any.whl", hash = "sha256:c647aa4a12dfbad9333ca4e71fe62ddc36f4e63b2d260a37a8b83d2f043ac309", size = 67548, upload-time = "2026-03-19T14:22:23.645Z" },
 ]
 
+[[package]]
+name = "azure-ai-agentserver-core"
+version = "2.0.0a20260331006"
+source = { registry = "https://pkgs.dev.azure.com/azure-sdk/public/_packaging/azure-sdk-for-python/pypi/simple/" }
+dependencies = [
+    { name = "hypercorn", marker = "sys_platform == 'darwin' or sys_platform == 'linux' or sys_platform == 'win32'" },
+    { name = "starlette", marker = "sys_platform == 'darwin' or sys_platform == 'linux' or sys_platform == 'win32'" },
+]
+sdist = { url = "https://pkgs.dev.azure.com/azure-sdk/29ec6040-b234-4e31-b139-33dc4287b756/_packaging/3572dbf9-b5ef-433b-9137-fc4d7768e7cc/pypi/download/azure-ai-agentserver-core/2a20260331006/azure_ai_agentserver_core-2.0.0a20260331006.tar.gz", hash = "sha256:97ec317f1c1266eb60d82ae02827220b2453fcabf2d7698c0ffefac8ade20a2f" }
+wheels = [
+    { url = "https://pkgs.dev.azure.com/azure-sdk/29ec6040-b234-4e31-b139-33dc4287b756/_packaging/3572dbf9-b5ef-433b-9137-fc4d7768e7cc/pypi/download/azure-ai-agentserver-core/2a20260331006/azure_ai_agentserver_core-2.0.0a20260331006-py3-none-any.whl", hash = "sha256:f475bf8101a41eef9d3189b3376842ece9add6277394fdbe78e74e5960c80c34" },
+]
+
+[[package]]
+name = "azure-ai-agentserver-invocations"
+version = "1.0.0a20260331006"
+source = { registry = "https://pkgs.dev.azure.com/azure-sdk/public/_packaging/azure-sdk-for-python/pypi/simple/" }
+dependencies = [
+    { name = "azure-ai-agentserver-core", marker = "sys_platform == 'darwin' or sys_platform == 'linux' or sys_platform == 'win32'" },
+]
+sdist = { url = "https://pkgs.dev.azure.com/azure-sdk/29ec6040-b234-4e31-b139-33dc4287b756/_packaging/3572dbf9-b5ef-433b-9137-fc4d7768e7cc/pypi/download/azure-ai-agentserver-invocations/1a20260331006/azure_ai_agentserver_invocations-1.0.0a20260331006.tar.gz", hash = "sha256:dd4d511a24388c79b937f960590569495076b889272b780746ef38c267e4f645" }
+wheels = [
+    { url = "https://pkgs.dev.azure.com/azure-sdk/29ec6040-b234-4e31-b139-33dc4287b756/_packaging/3572dbf9-b5ef-433b-9137-fc4d7768e7cc/pypi/download/azure-ai-agentserver-invocations/1a20260331006/azure_ai_agentserver_invocations-1.0.0a20260331006-py3-none-any.whl", hash = "sha256:615b49b73c448ea6c6c5f26e8be719d78e50a8f432fc89d4cff27397bd0297f0" },
+]
+
+[[package]]
+name = "azure-ai-agentserver-responses"
+version = "1.0.0a20260331006"
+source = { registry = "https://pkgs.dev.azure.com/azure-sdk/public/_packaging/azure-sdk-for-python/pypi/simple/" }
+dependencies = [
+    { name = "azure-ai-agentserver-core", marker = "sys_platform == 'darwin' or sys_platform == 'linux' or sys_platform == 'win32'" },
+    { name = "azure-core", marker = "sys_platform == 'darwin' or sys_platform == 'linux' or sys_platform == 'win32'" },
+    { name = "httpx", marker = "sys_platform == 'darwin' or sys_platform == 'linux' or sys_platform == 'win32'" },
+    { name = "isodate", marker = "sys_platform == 'darwin' or sys_platform == 'linux' or sys_platform == 'win32'" },
+]
+sdist = { url = "https://pkgs.dev.azure.com/azure-sdk/29ec6040-b234-4e31-b139-33dc4287b756/_packaging/3572dbf9-b5ef-433b-9137-fc4d7768e7cc/pypi/download/azure-ai-agentserver-responses/1a20260331006/azure_ai_agentserver_responses-1.0.0a20260331006.tar.gz", hash = "sha256:86f390c622697b5977d090cae182c929d4221abd5ddaf4838f9e0d3e0b4a2b2b" }
+wheels = [
+    { url = "https://pkgs.dev.azure.com/azure-sdk/29ec6040-b234-4e31-b139-33dc4287b756/_packaging/3572dbf9-b5ef-433b-9137-fc4d7768e7cc/pypi/download/azure-ai-agentserver-responses/1a20260331006/azure_ai_agentserver_responses-1.0.0a20260331006-py3-none-any.whl", hash = "sha256:ce84c3d6ce7a68c8ffcaee37bef45dc43199a66aa50fb876efc52d45e6a92925" },
+]
+
 [[package]]
 name = "azure-ai-inference"
 version = "1.0.0b9"
@@ -1423,7 +1485,7 @@ name = "clr-loader"
 version = "0.2.10"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
-    { name = "cffi", marker = "sys_platform == 'darwin' or sys_platform == 'linux' or sys_platform == 'win32'" },
+    { name = "cffi", marker = "(python_full_version < '3.14' and sys_platform == 'darwin') or (python_full_version < '3.14' and sys_platform == 'linux') or (python_full_version < '3.14' and sys_platform == 'win32')" },
 ]
 sdist = { url = "https://files.pythonhosted.org/packages/18/24/c12faf3f61614b3131b5c98d3bf0d376b49c7feaa73edca559aeb2aee080/clr_loader-0.2.10.tar.gz", hash = "sha256:81f114afbc5005bafc5efe5af1341d400e22137e275b042a8979f3feb9fc9446", size = 83605, upload-time = "2026-01-03T23:13:06.984Z" }
 wheels = [
@@ -2668,6 +2730,25 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/a9/ae/8a3a16ea4d202cb641b51d2681bdd3d482c1c592d7570b3fa264730829ce/huggingface_hub-1.8.0-py3-none-any.whl", hash = "sha256:d3eb5047bd4e33c987429de6020d4810d38a5bef95b3b40df9b17346b7f353f2", size = 625208, upload-time = "2026-03-25T16:01:26.603Z" },
 ]
 
+[[package]]
+name = "hypercorn"
+version = "0.18.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "exceptiongroup", marker = "(python_full_version < '3.11' and sys_platform == 'darwin') or (python_full_version < '3.11' and sys_platform == 'linux') or (python_full_version < '3.11' and sys_platform == 'win32')" },
+    { name = "h11", marker = "sys_platform == 'darwin' or sys_platform == 'linux' or sys_platform == 'win32'" },
+    { name = "h2", marker = "sys_platform == 'darwin' or sys_platform == 'linux' or sys_platform == 'win32'" },
+    { name = "priority", marker = "sys_platform == 'darwin' or sys_platform == 'linux' or sys_platform == 'win32'" },
+    { name = "taskgroup", marker = "(python_full_version < '3.11' and sys_platform == 'darwin') or (python_full_version < '3.11' and sys_platform == 'linux') or (python_full_version < '3.11' and sys_platform == 'win32')" },
+    { name = "tomli", marker = "(python_full_version < '3.11' and sys_platform == 'darwin') or (python_full_version < '3.11' and sys_platform == 'linux') or (python_full_version < '3.11' and sys_platform == 'win32')" },
+    { name = "typing-extensions", marker = "(python_full_version < '3.11' and sys_platform == 'darwin') or (python_full_version < '3.11' and sys_platform == 'linux') or (python_full_version < '3.11' and sys_platform == 'win32')" },
+    { name = "wsproto", marker = "sys_platform == 'darwin' or sys_platform == 'linux' or sys_platform == 'win32'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/44/01/39f41a014b83dd5c795217362f2ca9071cf243e6a75bdcd6cd5b944658cc/hypercorn-0.18.0.tar.gz", hash = "sha256:d63267548939c46b0247dc8e5b45a9947590e35e64ee73a23c074aa3cf88e9da", size = 68420, upload-time = "2025-11-08T13:54:04.78Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/93/35/850277d1b17b206bd10874c8a9a3f52e059452fb49bb0d22cbb908f6038b/hypercorn-0.18.0-py3-none-any.whl", hash = "sha256:225e268f2c1c2f28f6d8f6db8f40cb8c992963610c5725e13ccfcddccb24b1cd", size = 61640, upload-time = "2025-11-08T13:54:03.202Z" },
+]
+
 [[package]]
 name = "hyperframe"
 version = "6.1.0"
@@ -4673,8 +4754,8 @@ name = "powerfx"
 version = "0.0.34"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
-    { name = "cffi", marker = "sys_platform == 'darwin' or sys_platform == 'linux' or sys_platform == 'win32'" },
-    { name = "pythonnet", marker = "sys_platform == 'darwin' or sys_platform == 'linux' or sys_platform == 'win32'" },
+    { name = "cffi", marker = "(python_full_version < '3.14' and sys_platform == 'darwin') or (python_full_version < '3.14' and sys_platform == 'linux') or (python_full_version < '3.14' and sys_platform == 'win32')" },
+    { name = "pythonnet", marker = "(python_full_version < '3.14' and sys_platform == 'darwin') or (python_full_version < '3.14' and sys_platform == 'linux') or (python_full_version < '3.14' and sys_platform == 'win32')" },
 ]
 sdist = { url = "https://files.pythonhosted.org/packages/9f/fb/6c4bf87e0c74ca1c563921ce89ca1c5785b7576bca932f7255cdf81082a7/powerfx-0.0.34.tar.gz", hash = "sha256:956992e7afd272657ed16d80f4cad24ec95d9e4a79fb9dfa4a068a09e136af32", size = 3237555, upload-time = "2025-12-22T15:50:59.682Z" }
 wheels = [
@@ -4705,6 +4786,15 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/53/05/9cca1708bb8c65264124eb4b04251e0f65ce5bfc707080bb6b492d5a0df7/prek-0.3.8-py3-none-win_arm64.whl", hash = "sha256:a2614647aeafa817a5802ccb9561e92eedc20dcf840639a1b00826e2c2442515", size = 5190872, upload-time = "2026-03-23T08:23:29.463Z" },
 ]
 
+[[package]]
+name = "priority"
+version = "2.0.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/f5/3c/eb7c35f4dcede96fca1842dac5f4f5d15511aa4b52f3a961219e68ae9204/priority-2.0.0.tar.gz", hash = "sha256:c965d54f1b8d0d0b19479db3924c7c36cf672dbf2aec92d43fbdaf4492ba18c0", size = 24792, upload-time = "2021-06-27T10:15:05.487Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/5e/5f/82c8074f7e84978129347c2c6ec8b6c59f3584ff1a20bc3c940a3e061790/priority-2.0.0-py3-none-any.whl", hash = "sha256:6f8eefce5f3ad59baf2c080a664037bb4725cd0a790d53d59ab4059288faf6aa", size = 8946, upload-time = "2021-06-27T10:15:03.856Z" },
+]
+
 [[package]]
 name = "propcache"
 version = "0.4.1"
@@ -5341,7 +5431,7 @@ name = "pythonnet"
 version = "3.0.5"
 source = { registry = "https://pypi.org/simple" }
 dependencies = [
-    { name = "clr-loader", marker = "sys_platform == 'darwin' or sys_platform == 'linux' or sys_platform == 'win32'" },
+    { name = "clr-loader", marker = "(python_full_version < '3.14' and sys_platform == 'darwin') or (python_full_version < '3.14' and sys_platform == 'linux') or (python_full_version < '3.14' and sys_platform == 'win32')" },
 ]
 sdist = { url = "https://files.pythonhosted.org/packages/9a/d6/1afd75edd932306ae9bd2c2d961d603dc2b52fcec51b04afea464f1f6646/pythonnet-3.0.5.tar.gz", hash = "sha256:48e43ca463941b3608b32b4e236db92d8d40db4c58a75ace902985f76dac21cf", size = 239212, upload-time = "2024-12-13T08:30:44.393Z" }
 wheels = [
@@ -6351,6 +6441,19 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/99/55/db07de81b5c630da5cbf5c7df646580ca26dfaefa593667fc6f2fe016d2e/tabulate-0.10.0-py3-none-any.whl", hash = "sha256:f0b0622e567335c8fabaaa659f1b33bcb6ddfe2e496071b743aa113f8774f2d3", size = 39814, upload-time = "2026-03-04T18:55:31.284Z" },
 ]
 
+[[package]]
+name = "taskgroup"
+version = "0.2.2"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "exceptiongroup", marker = "(python_full_version < '3.11' and sys_platform == 'darwin') or (python_full_version < '3.11' and sys_platform == 'linux') or (python_full_version < '3.11' and sys_platform == 'win32')" },
+    { name = "typing-extensions", marker = "(python_full_version < '3.11' and sys_platform == 'darwin') or (python_full_version < '3.11' and sys_platform == 'linux') or (python_full_version < '3.11' and sys_platform == 'win32')" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/f0/8d/e218e0160cc1b692e6e0e5ba34e8865dbb171efeb5fc9a704544b3020605/taskgroup-0.2.2.tar.gz", hash = "sha256:078483ac3e78f2e3f973e2edbf6941374fbea81b9c5d0a96f51d297717f4752d", size = 11504, upload-time = "2025-01-03T09:24:13.761Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/d1/b1/74babcc824a57904e919f3af16d86c08b524c0691504baf038ef2d7f655c/taskgroup-0.2.2-py2.py3-none-any.whl", hash = "sha256:e2c53121609f4ae97303e9ea1524304b4de6faf9eb2c9280c7f87976479a52fb", size = 14237, upload-time = "2025-01-03T09:24:11.41Z" },
+]
+
 [[package]]
 name = "tau2"
 version = "0.0.1"
@@ -7050,6 +7153,18 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/1f/f6/a933bd70f98e9cf3e08167fc5cd7aaaca49147e48411c0bd5ae701bb2194/wrapt-1.17.3-py3-none-any.whl", hash = "sha256:7171ae35d2c33d326ac19dd8facb1e82e5fd04ef8c6c0e394d7af55a55051c22", size = 23591, upload-time = "2025-08-12T05:53:20.674Z" },
 ]
 
+[[package]]
+name = "wsproto"
+version = "1.3.2"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "h11", marker = "sys_platform == 'darwin' or sys_platform == 'linux' or sys_platform == 'win32'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/c7/79/12135bdf8b9c9367b8701c2c19a14c913c120b882d50b014ca0d38083c2c/wsproto-1.3.2.tar.gz", hash = "sha256:b86885dcf294e15204919950f666e06ffc6c7c114ca900b060d6e16293528294", size = 50116, upload-time = "2025-11-20T18:18:01.871Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/a4/f5/10b68b7b1544245097b2a1b8238f66f2fc6dcaeb24ba5d917f52bd2eed4f/wsproto-1.3.2-py3-none-any.whl", hash = "sha256:61eea322cdf56e8cc904bd3ad7573359a242ba65688716b0710a5eb12beab584", size = 24405, upload-time = "2025-11-20T18:18:00.454Z" },
+]
+
 [[package]]
 name = "yarl"
 version = "1.23.0"

From 951f61abb6ab8edbd6616f4065fda90bd99e52f5 Mon Sep 17 00:00:00 2001
From: Tao Chen <taochen@microsoft.com>
Date: Wed, 8 Apr 2026 10:38:50 -0700
Subject: [PATCH 02/13] Update dependency and add non streaming

---
 .../core/agent_framework/foundry/__init__.py  |  2 +-
 .../core/agent_framework/foundry/__init__.pyi |  4 +-
 .../__init__.py                               |  4 +-
 .../_responses.py                             | 75 ++++++++++++-----
 .../_shared.py                                |  4 +-
 .../packages/foundry_hosting/pyproject.toml   |  6 +-
 python/pyproject.toml                         | 11 +--
 .../foundry_vnext/helloworld/main.py          |  5 +-
 python/uv.lock                                | 81 ++++++++++++++-----
 9 files changed, 128 insertions(+), 64 deletions(-)

diff --git a/python/packages/core/agent_framework/foundry/__init__.py b/python/packages/core/agent_framework/foundry/__init__.py
index 8edaed5485..f438b2ab10 100644
--- a/python/packages/core/agent_framework/foundry/__init__.py
+++ b/python/packages/core/agent_framework/foundry/__init__.py
@@ -32,7 +32,7 @@
     "RawFoundryEmbeddingClient": ("agent_framework_foundry", "agent-framework-foundry"),
     "evaluate_foundry_target": ("agent_framework_foundry", "agent-framework-foundry"),
     "evaluate_traces": ("agent_framework_foundry", "agent-framework-foundry"),
-    "ResponsesHost": ("agent_framework_foundry_hosting", "agent-framework-foundry-hosting"),
+    "ResponsesHostServer": ("agent_framework_foundry_hosting", "agent-framework-foundry-hosting"),
 }
 
 
diff --git a/python/packages/core/agent_framework/foundry/__init__.pyi b/python/packages/core/agent_framework/foundry/__init__.pyi
index 133378ea2a..2d2a5e2920 100644
--- a/python/packages/core/agent_framework/foundry/__init__.pyi
+++ b/python/packages/core/agent_framework/foundry/__init__.pyi
@@ -20,7 +20,7 @@ from agent_framework_foundry import (
     evaluate_foundry_target,
     evaluate_traces,
 )
-from agent_framework_foundry_hosting import ResponsesHost
+from agent_framework_foundry_hosting import ResponsesHostServer
 from agent_framework_foundry_local import (
     FoundryLocalChatOptions,
     FoundryLocalClient,
@@ -45,7 +45,7 @@ __all__ = [
     "RawFoundryAgentChatClient",
     "RawFoundryChatClient",
     "RawFoundryEmbeddingClient",
-    "ResponsesHost",
+    "ResponsesHostServer",
     "evaluate_foundry_target",
     "evaluate_traces",
 ]
diff --git a/python/packages/foundry_hosting/agent_framework_foundry_hosting/__init__.py b/python/packages/foundry_hosting/agent_framework_foundry_hosting/__init__.py
index 595bdc8e1b..bad5e56fe0 100644
--- a/python/packages/foundry_hosting/agent_framework_foundry_hosting/__init__.py
+++ b/python/packages/foundry_hosting/agent_framework_foundry_hosting/__init__.py
@@ -2,11 +2,11 @@
 
 import importlib.metadata
 
-from ._responses import ResponsesHost
+from ._responses import ResponsesHostServer
 
 try:
     __version__ = importlib.metadata.version(__name__)
 except importlib.metadata.PackageNotFoundError:
     __version__ = "0.0.0"
 
-__all__ = ["ResponsesHost"]
+__all__ = ["ResponsesHostServer"]
diff --git a/python/packages/foundry_hosting/agent_framework_foundry_hosting/_responses.py b/python/packages/foundry_hosting/agent_framework_foundry_hosting/_responses.py
index 6e3afff5bc..1b962b63e0 100644
--- a/python/packages/foundry_hosting/agent_framework_foundry_hosting/_responses.py
+++ b/python/packages/foundry_hosting/agent_framework_foundry_hosting/_responses.py
@@ -4,9 +4,13 @@
 from collections.abc import AsyncIterable
 
 from agent_framework import Agent, HistoryProvider, Message
-from azure.ai.agentserver.core import AgentHost
-from azure.ai.agentserver.responses import ResponseContext, ResponseEventStream
-from azure.ai.agentserver.responses.hosting import ResponseHandler
+from azure.ai.agentserver.responses import (
+    ResponseContext,
+    ResponseEventStream,
+    ResponseProviderProtocol,
+    ResponsesServerOptions,
+)
+from azure.ai.agentserver.responses.hosting import ResponsesAgentServerHost
 from azure.ai.agentserver.responses.models import CreateResponse, get_input_text
 from typing_extensions import Any, Sequence
 
@@ -14,7 +18,14 @@
 
 
 class ResponsesHostContextProvider(HistoryProvider):
+    """A history provider that retrieves messages from a ResponseContext."""
+
     def __init__(self, context: ResponseContext):
+        """Initialize a ResponsesHostContextProvider.
+
+        Args:
+            context: The ResponseContext to retrieve messages from.
+        """
         super().__init__("responses-host", load_messages=True)
         self.context = context
 
@@ -25,7 +36,7 @@ async def get_messages(
         state: dict[str, Any] | None = None,
         **kwargs: Any,
     ) -> list[Message]:
-        history = await self.context.get_history_async()
+        history = await self.context.get_history()
         return to_messages(history)
 
     async def save_messages(
@@ -39,20 +50,36 @@ async def save_messages(
         pass
 
 
-class ResponsesHost(AgentHost):
-    def __init__(self, agent: Agent, **kwargs: Any) -> None:
-        application_insights_connection_string = kwargs.pop("application_insights_connection_string", None)
-        graceful_shutdown_timeout = kwargs.pop("graceful_shutdown_timeout", None)
-        log_level = kwargs.pop("log_level", None)
-        super().__init__(
-            application_insights_connection_string=application_insights_connection_string,
-            graceful_shutdown_timeout=graceful_shutdown_timeout,
-            log_level=log_level,
-        )
+class ResponsesHostServer(ResponsesAgentServerHost):
+    """A responses server host for an agent."""
+
+    def __init__(
+        self,
+        agent: Agent,
+        *,
+        prefix: str = "",
+        options: ResponsesServerOptions | None = None,
+        provider: ResponseProviderProtocol | None = None,
+        **kwargs: Any,
+    ) -> None:
+        """Initialize a ResponsesHostServer.
+
+        Args:
+            agent: The agent to handle responses for.
+            prefix: The URL prefix for the server.
+            options: Optional server options.
+            provider: Optional response provider.
+            **kwargs: Additional keyword arguments.
+
+        Note:
+            If the agent has a history provider with `load_messages=True`, it will be
+            replaced with a `ResponsesHostContextProvider` that will retrieve history
+            from the hosting infrastructure.
+        """
+        super().__init__(prefix=prefix, options=options, provider=provider, **kwargs)
 
         self.agent = agent
-        self.response_handler = ResponseHandler(self)
-        self.response_handler.create_handler(self._handle_create)  # type: ignore
+        self.create_handler(self._handle_create)  # pyright: ignore[reportUnknownMemberType]
 
     async def _handle_create(
         self,
@@ -74,17 +101,21 @@ async def _handle_create(
         else:
             self.agent.context_providers[history_provider_idx[0]] = ResponsesHostContextProvider(context)
 
-        stream = ResponseEventStream(
-            response_id=context.response_id,
-            model=getattr(request, "model", None),
-        )
+        input_items = get_input_text(request)
+
+        stream = ResponseEventStream(response_id=context.response_id, model=request.model)
 
         yield stream.emit_created()
         yield stream.emit_in_progress()
 
-        input_items = get_input_text(request)
+        if request.stream is None or request.stream is False:
+            # Run the agent in non-streaming mode
+            response = await self.agent.run(input_items, stream=False)
+            for item in stream.output_item_message(response.text):
+                yield item
+            yield stream.emit_completed()
 
-        # Start the response
+        # Start the streaming response
         message_item = stream.add_output_item_message()
         yield message_item.emit_added()
         text_content = message_item.add_text_content()
diff --git a/python/packages/foundry_hosting/agent_framework_foundry_hosting/_shared.py b/python/packages/foundry_hosting/agent_framework_foundry_hosting/_shared.py
index 3cd64bed1d..ec6d732915 100644
--- a/python/packages/foundry_hosting/agent_framework_foundry_hosting/_shared.py
+++ b/python/packages/foundry_hosting/agent_framework_foundry_hosting/_shared.py
@@ -4,6 +4,7 @@
 from azure.ai.agentserver.responses.models import (
     ComputerScreenshotContent,
     CreateResponse,
+    FunctionCallOutputItemParam,
     MessageContent,
     MessageContentInputFileContent,
     MessageContentInputImageContent,
@@ -13,7 +14,6 @@
     MessageContentRefusalContent,
     OutputItem,
     OutputItemFunctionToolCall,
-    OutputItemFunctionToolCallOutput,
     OutputItemMessage,
     OutputItemOutputMessage,
     OutputItemReasoningItem,
@@ -92,7 +92,7 @@ def _to_message(item: OutputItem) -> Message:
         )
 
     if item.type == "function_call_output":
-        fco = cast(OutputItemFunctionToolCallOutput, item)
+        fco = cast(FunctionCallOutputItemParam, item)
         output = fco.output if isinstance(fco.output, str) else str(fco.output)
         return Message(
             role="tool",
diff --git a/python/packages/foundry_hosting/pyproject.toml b/python/packages/foundry_hosting/pyproject.toml
index bce8bbfd69..1f0c42dbbf 100644
--- a/python/packages/foundry_hosting/pyproject.toml
+++ b/python/packages/foundry_hosting/pyproject.toml
@@ -24,9 +24,9 @@ classifiers = [
 ]
 dependencies = [
     "agent-framework-core>=1.0.0,<2",
-    "azure-ai-agentserver-core==2.0.0a20260331006",
-    "azure-ai-agentserver-responses==1.0.0a20260331006",
-    "azure-ai-agentserver-invocations==1.0.0a20260331006"
+    "azure-ai-agentserver-core",
+    "azure-ai-agentserver-responses",
+    "azure-ai-agentserver-invocations"
 ]
 
 [tool.uv]
diff --git a/python/pyproject.toml b/python/pyproject.toml
index c2b692e23c..4accd6d888 100644
--- a/python/pyproject.toml
+++ b/python/pyproject.toml
@@ -89,9 +89,9 @@ agent-framework-redis = { workspace = true }
 agent-framework-github-copilot = { workspace = true }
 agent-framework-claude = { workspace = true }
 agent-framework-orchestrations = { workspace = true }
-azure-ai-agentserver-responses = { index = "azure-sdk-feed" }
-azure-ai-agentserver-invocations = { index = "azure-sdk-feed" }
-azure-ai-agentserver-core = { index = "azure-sdk-feed" }
+azure-ai-agentserver-responses = { git = "https://github.com/Azure/azure-sdk-for-python.git", branch = "agentserver/responses", subdirectory = "sdk/agentserver/azure-ai-agentserver-responses" }
+azure-ai-agentserver-invocations = { git = "https://github.com/Azure/azure-sdk-for-python.git", branch = "agentserver/responses", subdirectory = "sdk/agentserver/azure-ai-agentserver-invocations" }
+azure-ai-agentserver-core = { git = "https://github.com/Azure/azure-sdk-for-python.git", branch = "agentserver/responses", subdirectory = "sdk/agentserver/azure-ai-agentserver-core" }
 litellm = { url = "https://files.pythonhosted.org/packages/57/77/0c6eca2cb049793ddf8ce9cdcd5123a35666c4962514788c4fc90edf1d3b/litellm-1.82.1-py3-none-any.whl" }
 
 [tool.ruff]
@@ -422,11 +422,6 @@ url = "https://test.pypi.org/simple/"
 publish-url = "https://test.pypi.org/legacy/"
 explicit = true
 
-[[tool.uv.index]]
-name = "azure-sdk-feed"
-url = "https://pkgs.dev.azure.com/azure-sdk/public/_packaging/azure-sdk-for-python/pypi/simple/"
-explicit = true
-
 [tool.flit.module]
 name = "agent_framework_meta"
 
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext/helloworld/main.py b/python/samples/05-end-to-end/hosted_agents/foundry_vnext/helloworld/main.py
index f1e1972393..6b63158095 100644
--- a/python/samples/05-end-to-end/hosted_agents/foundry_vnext/helloworld/main.py
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_vnext/helloworld/main.py
@@ -4,7 +4,8 @@
 import os
 
 from agent_framework import Agent
-from agent_framework.foundry import FoundryChatClient, ResponsesHost
+from agent_framework.foundry import FoundryChatClient, ResponsesHostServer
+from azure.ai.agentserver.responses import InMemoryResponseProvider
 from azure.identity import AzureCliCredential
 from dotenv import load_dotenv
 
@@ -25,7 +26,7 @@ def main():
         instructions="You are a friendly assistant. Keep your answers brief.",
     )
 
-    server = ResponsesHost(agent)
+    server = ResponsesHostServer(agent, provider=InMemoryResponseProvider())
     server.run()
 
 
diff --git a/python/uv.lock b/python/uv.lock
index a713525a20..b9cf525334 100644
--- a/python/uv.lock
+++ b/python/uv.lock
@@ -514,9 +514,9 @@ dependencies = [
 [package.metadata]
 requires-dist = [
     { name = "agent-framework-core", editable = "packages/core" },
-    { name = "azure-ai-agentserver-core", specifier = "==2.0.0a20260331006", index = "https://pkgs.dev.azure.com/azure-sdk/public/_packaging/azure-sdk-for-python/pypi/simple/" },
-    { name = "azure-ai-agentserver-invocations", specifier = "==1.0.0a20260331006", index = "https://pkgs.dev.azure.com/azure-sdk/public/_packaging/azure-sdk-for-python/pypi/simple/" },
-    { name = "azure-ai-agentserver-responses", specifier = "==1.0.0a20260331006", index = "https://pkgs.dev.azure.com/azure-sdk/public/_packaging/azure-sdk-for-python/pypi/simple/" },
+    { name = "azure-ai-agentserver-core", git = "https://github.com/Azure/azure-sdk-for-python.git?subdirectory=sdk%2Fagentserver%2Fazure-ai-agentserver-core&branch=agentserver%2Fresponses" },
+    { name = "azure-ai-agentserver-invocations", git = "https://github.com/Azure/azure-sdk-for-python.git?subdirectory=sdk%2Fagentserver%2Fazure-ai-agentserver-invocations&branch=agentserver%2Fresponses" },
+    { name = "azure-ai-agentserver-responses", git = "https://github.com/Azure/azure-sdk-for-python.git?subdirectory=sdk%2Fagentserver%2Fazure-ai-agentserver-responses&branch=agentserver%2Fresponses" },
 ]
 
 [[package]]
@@ -1020,43 +1020,34 @@ wheels = [
 
 [[package]]
 name = "azure-ai-agentserver-core"
-version = "2.0.0a20260331006"
-source = { registry = "https://pkgs.dev.azure.com/azure-sdk/public/_packaging/azure-sdk-for-python/pypi/simple/" }
+version = "2.0.0b1"
+source = { git = "https://github.com/Azure/azure-sdk-for-python.git?subdirectory=sdk%2Fagentserver%2Fazure-ai-agentserver-core&branch=agentserver%2Fresponses#43579f686f51ebed23b066d06c90a544c0070a0b" }
 dependencies = [
+    { name = "azure-monitor-opentelemetry-exporter", marker = "sys_platform == 'darwin' or sys_platform == 'linux' or sys_platform == 'win32'" },
     { name = "hypercorn", marker = "sys_platform == 'darwin' or sys_platform == 'linux' or sys_platform == 'win32'" },
+    { name = "opentelemetry-api", marker = "sys_platform == 'darwin' or sys_platform == 'linux' or sys_platform == 'win32'" },
+    { name = "opentelemetry-exporter-otlp-proto-grpc", marker = "sys_platform == 'darwin' or sys_platform == 'linux' or sys_platform == 'win32'" },
+    { name = "opentelemetry-sdk", marker = "sys_platform == 'darwin' or sys_platform == 'linux' or sys_platform == 'win32'" },
     { name = "starlette", marker = "sys_platform == 'darwin' or sys_platform == 'linux' or sys_platform == 'win32'" },
 ]
-sdist = { url = "https://pkgs.dev.azure.com/azure-sdk/29ec6040-b234-4e31-b139-33dc4287b756/_packaging/3572dbf9-b5ef-433b-9137-fc4d7768e7cc/pypi/download/azure-ai-agentserver-core/2a20260331006/azure_ai_agentserver_core-2.0.0a20260331006.tar.gz", hash = "sha256:97ec317f1c1266eb60d82ae02827220b2453fcabf2d7698c0ffefac8ade20a2f" }
-wheels = [
-    { url = "https://pkgs.dev.azure.com/azure-sdk/29ec6040-b234-4e31-b139-33dc4287b756/_packaging/3572dbf9-b5ef-433b-9137-fc4d7768e7cc/pypi/download/azure-ai-agentserver-core/2a20260331006/azure_ai_agentserver_core-2.0.0a20260331006-py3-none-any.whl", hash = "sha256:f475bf8101a41eef9d3189b3376842ece9add6277394fdbe78e74e5960c80c34" },
-]
 
 [[package]]
 name = "azure-ai-agentserver-invocations"
-version = "1.0.0a20260331006"
-source = { registry = "https://pkgs.dev.azure.com/azure-sdk/public/_packaging/azure-sdk-for-python/pypi/simple/" }
+version = "1.0.0b1"
+source = { git = "https://github.com/Azure/azure-sdk-for-python.git?subdirectory=sdk%2Fagentserver%2Fazure-ai-agentserver-invocations&branch=agentserver%2Fresponses#43579f686f51ebed23b066d06c90a544c0070a0b" }
 dependencies = [
     { name = "azure-ai-agentserver-core", marker = "sys_platform == 'darwin' or sys_platform == 'linux' or sys_platform == 'win32'" },
 ]
-sdist = { url = "https://pkgs.dev.azure.com/azure-sdk/29ec6040-b234-4e31-b139-33dc4287b756/_packaging/3572dbf9-b5ef-433b-9137-fc4d7768e7cc/pypi/download/azure-ai-agentserver-invocations/1a20260331006/azure_ai_agentserver_invocations-1.0.0a20260331006.tar.gz", hash = "sha256:dd4d511a24388c79b937f960590569495076b889272b780746ef38c267e4f645" }
-wheels = [
-    { url = "https://pkgs.dev.azure.com/azure-sdk/29ec6040-b234-4e31-b139-33dc4287b756/_packaging/3572dbf9-b5ef-433b-9137-fc4d7768e7cc/pypi/download/azure-ai-agentserver-invocations/1a20260331006/azure_ai_agentserver_invocations-1.0.0a20260331006-py3-none-any.whl", hash = "sha256:615b49b73c448ea6c6c5f26e8be719d78e50a8f432fc89d4cff27397bd0297f0" },
-]
 
 [[package]]
 name = "azure-ai-agentserver-responses"
-version = "1.0.0a20260331006"
-source = { registry = "https://pkgs.dev.azure.com/azure-sdk/public/_packaging/azure-sdk-for-python/pypi/simple/" }
+version = "1.0.0b1"
+source = { git = "https://github.com/Azure/azure-sdk-for-python.git?subdirectory=sdk%2Fagentserver%2Fazure-ai-agentserver-responses&branch=agentserver%2Fresponses#43579f686f51ebed23b066d06c90a544c0070a0b" }
 dependencies = [
     { name = "azure-ai-agentserver-core", marker = "sys_platform == 'darwin' or sys_platform == 'linux' or sys_platform == 'win32'" },
     { name = "azure-core", marker = "sys_platform == 'darwin' or sys_platform == 'linux' or sys_platform == 'win32'" },
-    { name = "httpx", marker = "sys_platform == 'darwin' or sys_platform == 'linux' or sys_platform == 'win32'" },
     { name = "isodate", marker = "sys_platform == 'darwin' or sys_platform == 'linux' or sys_platform == 'win32'" },
 ]
-sdist = { url = "https://pkgs.dev.azure.com/azure-sdk/29ec6040-b234-4e31-b139-33dc4287b756/_packaging/3572dbf9-b5ef-433b-9137-fc4d7768e7cc/pypi/download/azure-ai-agentserver-responses/1a20260331006/azure_ai_agentserver_responses-1.0.0a20260331006.tar.gz", hash = "sha256:86f390c622697b5977d090cae182c929d4221abd5ddaf4838f9e0d3e0b4a2b2b" }
-wheels = [
-    { url = "https://pkgs.dev.azure.com/azure-sdk/29ec6040-b234-4e31-b139-33dc4287b756/_packaging/3572dbf9-b5ef-433b-9137-fc4d7768e7cc/pypi/download/azure-ai-agentserver-responses/1a20260331006/azure_ai_agentserver_responses-1.0.0a20260331006-py3-none-any.whl", hash = "sha256:ce84c3d6ce7a68c8ffcaee37bef45dc43199a66aa50fb876efc52d45e6a92925" },
-]
 
 [[package]]
 name = "azure-ai-inference"
@@ -1170,6 +1161,23 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/49/9a/417b3a533e01953a7c618884df2cb05a71e7b68bdbce4fbdb62349d2a2e8/azure_identity-1.25.3-py3-none-any.whl", hash = "sha256:f4d0b956a8146f30333e071374171f3cfa7bdb8073adb8c3814b65567aa7447c", size = 192138, upload-time = "2026-03-13T01:12:22.951Z" },
 ]
 
+[[package]]
+name = "azure-monitor-opentelemetry-exporter"
+version = "1.0.0b51"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "azure-core", marker = "sys_platform == 'darwin' or sys_platform == 'linux' or sys_platform == 'win32'" },
+    { name = "azure-identity", marker = "sys_platform == 'darwin' or sys_platform == 'linux' or sys_platform == 'win32'" },
+    { name = "msrest", marker = "sys_platform == 'darwin' or sys_platform == 'linux' or sys_platform == 'win32'" },
+    { name = "opentelemetry-api", marker = "sys_platform == 'darwin' or sys_platform == 'linux' or sys_platform == 'win32'" },
+    { name = "opentelemetry-sdk", marker = "sys_platform == 'darwin' or sys_platform == 'linux' or sys_platform == 'win32'" },
+    { name = "psutil", marker = "sys_platform == 'darwin' or sys_platform == 'linux' or sys_platform == 'win32'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/bc/a4/a6cd2d389bc1009300bcd57c9e2ace4b7e7ae1e5dc0bda415ee803629cf2/azure_monitor_opentelemetry_exporter-1.0.0b51.tar.gz", hash = "sha256:a6171c34326bcd6216938bb40d715c15f1f22984ac1986fc97231336d8ac4c3c", size = 319837, upload-time = "2026-04-06T21:45:46.378Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/ea/1a/6b0b7a6181b42709103a65a676c89fd5055cb1d1b281ebe10c49254a170f/azure_monitor_opentelemetry_exporter-1.0.0b51-py2.py3-none-any.whl", hash = "sha256:6572cac11f96e3b18ae1187cb35cf3b40d0004655dae8048896c41c765bea530", size = 242104, upload-time = "2026-04-06T21:45:47.856Z" },
+]
+
 [[package]]
 name = "azure-search-documents"
 version = "11.7.0b2"
@@ -3685,6 +3693,22 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/5e/75/bd9b7bb966668920f06b200e84454c8f3566b102183bc55c5473d96cb2b9/msal_extensions-1.3.1-py3-none-any.whl", hash = "sha256:96d3de4d034504e969ac5e85bae8106c8373b5c6568e4c8fa7af2eca9dbe6bca", size = 20583, upload-time = "2025-03-14T23:51:03.016Z" },
 ]
 
+[[package]]
+name = "msrest"
+version = "0.7.1"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "azure-core", marker = "sys_platform == 'darwin' or sys_platform == 'linux' or sys_platform == 'win32'" },
+    { name = "certifi", marker = "sys_platform == 'darwin' or sys_platform == 'linux' or sys_platform == 'win32'" },
+    { name = "isodate", marker = "sys_platform == 'darwin' or sys_platform == 'linux' or sys_platform == 'win32'" },
+    { name = "requests", marker = "sys_platform == 'darwin' or sys_platform == 'linux' or sys_platform == 'win32'" },
+    { name = "requests-oauthlib", marker = "sys_platform == 'darwin' or sys_platform == 'linux' or sys_platform == 'win32'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/68/77/8397c8fb8fc257d8ea0fa66f8068e073278c65f05acb17dcb22a02bfdc42/msrest-0.7.1.zip", hash = "sha256:6e7661f46f3afd88b75667b7187a92829924446c7ea1d169be8c4bb7eeb788b9", size = 175332, upload-time = "2022-06-13T22:41:25.111Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/15/cf/f2966a2638144491f8696c27320d5219f48a072715075d168b31d3237720/msrest-0.7.1-py3-none-any.whl", hash = "sha256:21120a810e1233e5e6cc7fe40b474eeb4ec6f757a15d7cf86702c369f9567c32", size = 85384, upload-time = "2022-06-13T22:41:22.42Z" },
+]
+
 [[package]]
 name = "multidict"
 version = "6.7.1"
@@ -5734,6 +5758,19 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/d7/8e/7540e8a2036f79a125c1d2ebadf69ed7901608859186c856fa0388ef4197/requests-2.33.1-py3-none-any.whl", hash = "sha256:4e6d1ef462f3626a1f0a0a9c42dd93c63bad33f9f1c1937509b8c5c8718ab56a", size = 64947, upload-time = "2026-03-30T16:09:13.83Z" },
 ]
 
+[[package]]
+name = "requests-oauthlib"
+version = "2.0.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "oauthlib", marker = "sys_platform == 'darwin' or sys_platform == 'linux' or sys_platform == 'win32'" },
+    { name = "requests", marker = "sys_platform == 'darwin' or sys_platform == 'linux' or sys_platform == 'win32'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/42/f2/05f29bc3913aea15eb670be136045bf5c5bbf4b99ecb839da9b422bb2c85/requests-oauthlib-2.0.0.tar.gz", hash = "sha256:b3dffaebd884d8cd778494369603a9e7b58d29111bf6b41bdc2dcd87203af4e9", size = 55650, upload-time = "2024-03-22T20:32:29.939Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/3b/5d/63d4ae3b9daea098d5d6f5da83984853c1bbacd5dc826764b249fe119d24/requests_oauthlib-2.0.0-py2.py3-none-any.whl", hash = "sha256:7dd8a5c40426b779b0868c404bdef9768deccf22749cde15852df527e6269b36", size = 24179, upload-time = "2024-03-22T20:32:28.055Z" },
+]
+
 [[package]]
 name = "rich"
 version = "13.9.4"

From 359a721cf60ffe8f2b993e36f43abfa18ca92617 Mon Sep 17 00:00:00 2001
From: Tao Chen <taochen@microsoft.com>
Date: Wed, 8 Apr 2026 11:19:15 -0700
Subject: [PATCH 03/13] Add more samples

---
 .../_responses.py                             | 16 ++---
 .../_shared.py                                | 24 +------
 .../helloworld/.dockerignore                  |  0
 .../helloworld/Dockerfile                     |  0
 .../helloworld/agent.manifest.yaml            |  0
 .../helloworld/agent.yaml                     |  0
 .../helloworld/main.py                        |  1 -
 .../helloworld/requirements.txt               |  0
 .../hosted_mcp_tools/.dockerignore            |  6 ++
 .../hosted_mcp_tools/Dockerfile               | 16 +++++
 .../hosted_mcp_tools/agent.manifest.yaml      | 16 +++++
 .../hosted_mcp_tools/agent.yaml               |  8 +++
 .../hosted_mcp_tools/main.py                  | 65 +++++++++++++++++++
 .../hosted_mcp_tools/requirements.txt         |  2 +
 .../local_tools/.dockerignore                 |  6 ++
 .../local_tools/Dockerfile                    | 16 +++++
 .../local_tools/agent.manifest.yaml           | 16 +++++
 .../local_tools/agent.yaml                    |  8 +++
 .../local_tools/main.py                       | 46 +++++++++++++
 .../local_tools/requirements.txt              |  2 +
 .../workflows/.dockerignore                   |  6 ++
 .../workflows/Dockerfile                      | 16 +++++
 .../workflows/agent.manifest.yaml             | 16 +++++
 .../workflows/agent.yaml                      |  8 +++
 .../foundry_vnext_responses/workflows/main.py | 49 ++++++++++++++
 .../workflows/requirements.txt                |  2 +
 26 files changed, 312 insertions(+), 33 deletions(-)
 rename python/samples/05-end-to-end/hosted_agents/{foundry_vnext => foundry_vnext_responses}/helloworld/.dockerignore (100%)
 rename python/samples/05-end-to-end/hosted_agents/{foundry_vnext => foundry_vnext_responses}/helloworld/Dockerfile (100%)
 rename python/samples/05-end-to-end/hosted_agents/{foundry_vnext => foundry_vnext_responses}/helloworld/agent.manifest.yaml (100%)
 rename python/samples/05-end-to-end/hosted_agents/{foundry_vnext => foundry_vnext_responses}/helloworld/agent.yaml (100%)
 rename python/samples/05-end-to-end/hosted_agents/{foundry_vnext => foundry_vnext_responses}/helloworld/main.py (99%)
 rename python/samples/05-end-to-end/hosted_agents/{foundry_vnext => foundry_vnext_responses}/helloworld/requirements.txt (100%)
 create mode 100644 python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/hosted_mcp_tools/.dockerignore
 create mode 100644 python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/hosted_mcp_tools/Dockerfile
 create mode 100644 python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/hosted_mcp_tools/agent.manifest.yaml
 create mode 100644 python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/hosted_mcp_tools/agent.yaml
 create mode 100644 python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/hosted_mcp_tools/main.py
 create mode 100644 python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/hosted_mcp_tools/requirements.txt
 create mode 100644 python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/local_tools/.dockerignore
 create mode 100644 python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/local_tools/Dockerfile
 create mode 100644 python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/local_tools/agent.manifest.yaml
 create mode 100644 python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/local_tools/agent.yaml
 create mode 100644 python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/local_tools/main.py
 create mode 100644 python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/local_tools/requirements.txt
 create mode 100644 python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/workflows/.dockerignore
 create mode 100644 python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/workflows/Dockerfile
 create mode 100644 python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/workflows/agent.manifest.yaml
 create mode 100644 python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/workflows/agent.yaml
 create mode 100644 python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/workflows/main.py
 create mode 100644 python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/workflows/requirements.txt

diff --git a/python/packages/foundry_hosting/agent_framework_foundry_hosting/_responses.py b/python/packages/foundry_hosting/agent_framework_foundry_hosting/_responses.py
index 1b962b63e0..eca938afbd 100644
--- a/python/packages/foundry_hosting/agent_framework_foundry_hosting/_responses.py
+++ b/python/packages/foundry_hosting/agent_framework_foundry_hosting/_responses.py
@@ -3,7 +3,7 @@
 import asyncio
 from collections.abc import AsyncIterable
 
-from agent_framework import Agent, HistoryProvider, Message
+from agent_framework import BaseAgent, HistoryProvider, Message, SupportsAgentRun
 from azure.ai.agentserver.responses import (
     ResponseContext,
     ResponseEventStream,
@@ -14,7 +14,7 @@
 from azure.ai.agentserver.responses.models import CreateResponse, get_input_text
 from typing_extensions import Any, Sequence
 
-from ._shared import extract_chat_options, to_messages
+from ._shared import to_messages
 
 
 class ResponsesHostContextProvider(HistoryProvider):
@@ -55,7 +55,7 @@ class ResponsesHostServer(ResponsesAgentServerHost):
 
     def __init__(
         self,
-        agent: Agent,
+        agent: BaseAgent,
         *,
         prefix: str = "",
         options: ResponsesServerOptions | None = None,
@@ -78,6 +78,9 @@ def __init__(
         """
         super().__init__(prefix=prefix, options=options, provider=provider, **kwargs)
 
+        if not isinstance(agent, SupportsAgentRun):
+            raise TypeError("Agent must support the SupportsAgentRun interface")
+
         self.agent = agent
         self.create_handler(self._handle_create)  # pyright: ignore[reportUnknownMemberType]
 
@@ -122,13 +125,8 @@ async def _handle_create(
         yield text_content.emit_added()
 
         # Invoke the MAF agent
-        chat_options = extract_chat_options(request)
         full_text = ""
-        async for update in self.agent.run(
-            input_items,
-            options=chat_options,
-            stream=True,
-        ):
+        async for update in self.agent.run(input_items, stream=True):
             full_text += update.text
             yield text_content.emit_delta(update.text)
 
diff --git a/python/packages/foundry_hosting/agent_framework_foundry_hosting/_shared.py b/python/packages/foundry_hosting/agent_framework_foundry_hosting/_shared.py
index ec6d732915..f76306af11 100644
--- a/python/packages/foundry_hosting/agent_framework_foundry_hosting/_shared.py
+++ b/python/packages/foundry_hosting/agent_framework_foundry_hosting/_shared.py
@@ -1,9 +1,8 @@
 # Copyright (c) Microsoft. All rights reserved.
 
-from agent_framework import ChatOptions, Content, Message
+from agent_framework import Content, Message
 from azure.ai.agentserver.responses.models import (
     ComputerScreenshotContent,
-    CreateResponse,
     FunctionCallOutputItemParam,
     MessageContent,
     MessageContentInputFileContent,
@@ -26,27 +25,6 @@
 from typing_extensions import Sequence, cast
 
 
-def extract_chat_options(request: CreateResponse) -> ChatOptions:
-    """Extracts chat options from a CreateResponse request.
-
-    Args:
-        request (CreateResponse): The CreateResponse request object containing the chat options.
-
-    Returns:
-        ChatOptions: The extracted chat options.
-    """
-    options = ChatOptions()
-
-    if request.temperature is not None:
-        options["temperature"] = request.temperature
-    if request.top_p is not None:
-        options["top_p"] = request.top_p
-    if request.max_output_tokens is not None:
-        options["max_tokens"] = request.max_output_tokens
-
-    return options
-
-
 def to_messages(history: Sequence[OutputItem]) -> list[Message]:
     """Converts a sequence of OutputItem objects to a list of Message objects.
 
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext/helloworld/.dockerignore b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/helloworld/.dockerignore
similarity index 100%
rename from python/samples/05-end-to-end/hosted_agents/foundry_vnext/helloworld/.dockerignore
rename to python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/helloworld/.dockerignore
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext/helloworld/Dockerfile b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/helloworld/Dockerfile
similarity index 100%
rename from python/samples/05-end-to-end/hosted_agents/foundry_vnext/helloworld/Dockerfile
rename to python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/helloworld/Dockerfile
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext/helloworld/agent.manifest.yaml b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/helloworld/agent.manifest.yaml
similarity index 100%
rename from python/samples/05-end-to-end/hosted_agents/foundry_vnext/helloworld/agent.manifest.yaml
rename to python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/helloworld/agent.manifest.yaml
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext/helloworld/agent.yaml b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/helloworld/agent.yaml
similarity index 100%
rename from python/samples/05-end-to-end/hosted_agents/foundry_vnext/helloworld/agent.yaml
rename to python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/helloworld/agent.yaml
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext/helloworld/main.py b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/helloworld/main.py
similarity index 99%
rename from python/samples/05-end-to-end/hosted_agents/foundry_vnext/helloworld/main.py
rename to python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/helloworld/main.py
index 6b63158095..7475b287f5 100644
--- a/python/samples/05-end-to-end/hosted_agents/foundry_vnext/helloworld/main.py
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/helloworld/main.py
@@ -1,6 +1,5 @@
 # Copyright (c) Microsoft. All rights reserved.
 
-
 import os
 
 from agent_framework import Agent
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext/helloworld/requirements.txt b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/helloworld/requirements.txt
similarity index 100%
rename from python/samples/05-end-to-end/hosted_agents/foundry_vnext/helloworld/requirements.txt
rename to python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/helloworld/requirements.txt
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/hosted_mcp_tools/.dockerignore b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/hosted_mcp_tools/.dockerignore
new file mode 100644
index 0000000000..008e6e6616
--- /dev/null
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/hosted_mcp_tools/.dockerignore
@@ -0,0 +1,6 @@
+.venv
+__pycache__
+*.pyc
+*.pyo
+*.pyd
+.Python
\ No newline at end of file
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/hosted_mcp_tools/Dockerfile b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/hosted_mcp_tools/Dockerfile
new file mode 100644
index 0000000000..eaffb94f19
--- /dev/null
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/hosted_mcp_tools/Dockerfile
@@ -0,0 +1,16 @@
+FROM python:3.12-slim
+
+WORKDIR /app
+
+COPY . user_agent/
+WORKDIR /app/user_agent
+
+RUN if [ -f requirements.txt ]; then \
+        pip install -r requirements.txt; \
+    else \
+        echo "No requirements.txt found"; \
+    fi
+
+EXPOSE 8088
+
+CMD ["python", "main.py"]
\ No newline at end of file
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/hosted_mcp_tools/agent.manifest.yaml b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/hosted_mcp_tools/agent.manifest.yaml
new file mode 100644
index 0000000000..ad074d6b89
--- /dev/null
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/hosted_mcp_tools/agent.manifest.yaml
@@ -0,0 +1,16 @@
+name: echo-agent-responses-streaming
+description: >
+  A simple echo agent that streams responses word-by-word using the
+  azure-ai-agentserver-responses SDK with Server-Sent Events (SSE).
+metadata:
+  tags:
+    - AI Agent Hosting
+    - Azure AI AgentServer
+    - Responses Protocol
+    - Streaming
+template:
+  name: echo-agent-responses-streaming
+  kind: hosted
+  protocols:
+    - protocol: responses
+      version: v0.1.0
\ No newline at end of file
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/hosted_mcp_tools/agent.yaml b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/hosted_mcp_tools/agent.yaml
new file mode 100644
index 0000000000..3738c690a1
--- /dev/null
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/hosted_mcp_tools/agent.yaml
@@ -0,0 +1,8 @@
+kind: hosted
+name: echo-agent-responses-streaming
+protocols:
+  - protocol: responses
+    version: v0.1.0
+resources:
+  cpu: "0.25"
+  memory: 0.5Gi
\ No newline at end of file
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/hosted_mcp_tools/main.py b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/hosted_mcp_tools/main.py
new file mode 100644
index 0000000000..164ef87fcf
--- /dev/null
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/hosted_mcp_tools/main.py
@@ -0,0 +1,65 @@
+# Copyright (c) Microsoft. All rights reserved.
+
+import os
+
+from agent_framework import Agent
+from agent_framework.foundry import FoundryChatClient, ResponsesHostServer
+from agent_framework.orchestrations import GroupChatBuilder, GroupChatState
+from azure.ai.agentserver.responses import InMemoryResponseProvider
+from azure.identity import AzureCliCredential
+from dotenv import load_dotenv
+
+# Load environment variables from .env file
+load_dotenv()
+
+
+def round_robin_selector(state: GroupChatState) -> str:
+    """A round-robin selector function that picks the next speaker based on the current round index."""
+
+    participant_names = list(state.participants.keys())
+    return participant_names[state.current_round % len(participant_names)]
+
+
+def main():
+    client = FoundryChatClient(
+        project_endpoint=os.environ["FOUNDRY_PROJECT_ENDPOINT"],
+        model=os.environ["FOUNDRY_MODEL"],
+        credential=AzureCliCredential(),
+    )
+
+    writer_agent = Agent(
+        client=client,
+        instructions=(
+            "You are an excellent content writer. You create new content and edit contents based on the feedback."
+        ),
+        name="writer",
+    )
+
+    reviewer_agent = Agent(
+        client=client,
+        instructions=(
+            "You are an excellent content reviewer."
+            "Provide actionable feedback to the writer about the provided content."
+            "Provide the feedback in the most concise manner possible."
+        ),
+        name="reviewer",
+    )
+
+    workflow_agent = (
+        GroupChatBuilder(
+            participants=[writer_agent, reviewer_agent],
+            # Set a hard termination condition to stop after 4 messages:
+            # User message + writer message + reviewer message + writer message
+            termination_condition=lambda conversation: len(conversation) >= 4,
+            selection_func=round_robin_selector,
+        )
+        .build()
+        .as_agent()
+    )
+
+    server = ResponsesHostServer(workflow_agent, provider=InMemoryResponseProvider())
+    server.run()
+
+
+if __name__ == "__main__":
+    main()
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/hosted_mcp_tools/requirements.txt b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/hosted_mcp_tools/requirements.txt
new file mode 100644
index 0000000000..f7dc62f3e3
--- /dev/null
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/hosted_mcp_tools/requirements.txt
@@ -0,0 +1,2 @@
+agent-framework
+agent-framework-foundry-hosting
\ No newline at end of file
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/local_tools/.dockerignore b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/local_tools/.dockerignore
new file mode 100644
index 0000000000..008e6e6616
--- /dev/null
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/local_tools/.dockerignore
@@ -0,0 +1,6 @@
+.venv
+__pycache__
+*.pyc
+*.pyo
+*.pyd
+.Python
\ No newline at end of file
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/local_tools/Dockerfile b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/local_tools/Dockerfile
new file mode 100644
index 0000000000..eaffb94f19
--- /dev/null
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/local_tools/Dockerfile
@@ -0,0 +1,16 @@
+FROM python:3.12-slim
+
+WORKDIR /app
+
+COPY . user_agent/
+WORKDIR /app/user_agent
+
+RUN if [ -f requirements.txt ]; then \
+        pip install -r requirements.txt; \
+    else \
+        echo "No requirements.txt found"; \
+    fi
+
+EXPOSE 8088
+
+CMD ["python", "main.py"]
\ No newline at end of file
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/local_tools/agent.manifest.yaml b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/local_tools/agent.manifest.yaml
new file mode 100644
index 0000000000..ad074d6b89
--- /dev/null
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/local_tools/agent.manifest.yaml
@@ -0,0 +1,16 @@
+name: echo-agent-responses-streaming
+description: >
+  A simple echo agent that streams responses word-by-word using the
+  azure-ai-agentserver-responses SDK with Server-Sent Events (SSE).
+metadata:
+  tags:
+    - AI Agent Hosting
+    - Azure AI AgentServer
+    - Responses Protocol
+    - Streaming
+template:
+  name: echo-agent-responses-streaming
+  kind: hosted
+  protocols:
+    - protocol: responses
+      version: v0.1.0
\ No newline at end of file
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/local_tools/agent.yaml b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/local_tools/agent.yaml
new file mode 100644
index 0000000000..3738c690a1
--- /dev/null
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/local_tools/agent.yaml
@@ -0,0 +1,8 @@
+kind: hosted
+name: echo-agent-responses-streaming
+protocols:
+  - protocol: responses
+    version: v0.1.0
+resources:
+  cpu: "0.25"
+  memory: 0.5Gi
\ No newline at end of file
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/local_tools/main.py b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/local_tools/main.py
new file mode 100644
index 0000000000..690f87d93c
--- /dev/null
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/local_tools/main.py
@@ -0,0 +1,46 @@
+# Copyright (c) Microsoft. All rights reserved.
+
+import os
+from random import randint
+
+from agent_framework import Agent, tool
+from agent_framework.foundry import FoundryChatClient, ResponsesHostServer
+from azure.ai.agentserver.responses import InMemoryResponseProvider
+from azure.identity import AzureCliCredential
+from dotenv import load_dotenv
+from pydantic import Field
+from typing_extensions import Annotated
+
+# Load environment variables from .env file
+load_dotenv()
+
+
+@tool(approval_mode="never_require")
+def get_weather(
+    location: Annotated[str, Field(description="The location to get the weather for.")],
+) -> str:
+    """Get the weather for a given location."""
+    conditions = ["sunny", "cloudy", "rainy", "stormy"]
+    return f"The weather in {location} is {conditions[randint(0, 3)]} with a high of {randint(10, 30)}°C."
+
+
+def main():
+    client = FoundryChatClient(
+        project_endpoint=os.environ["FOUNDRY_PROJECT_ENDPOINT"],
+        model=os.environ["FOUNDRY_MODEL"],
+        credential=AzureCliCredential(),
+    )
+
+    agent = Agent(
+        client=client,
+        name="HelloAgent",
+        instructions="You are a friendly assistant. Keep your answers brief.",
+        tools=[get_weather],
+    )
+
+    server = ResponsesHostServer(agent, provider=InMemoryResponseProvider())
+    server.run()
+
+
+if __name__ == "__main__":
+    main()
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/local_tools/requirements.txt b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/local_tools/requirements.txt
new file mode 100644
index 0000000000..f7dc62f3e3
--- /dev/null
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/local_tools/requirements.txt
@@ -0,0 +1,2 @@
+agent-framework
+agent-framework-foundry-hosting
\ No newline at end of file
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/workflows/.dockerignore b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/workflows/.dockerignore
new file mode 100644
index 0000000000..008e6e6616
--- /dev/null
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/workflows/.dockerignore
@@ -0,0 +1,6 @@
+.venv
+__pycache__
+*.pyc
+*.pyo
+*.pyd
+.Python
\ No newline at end of file
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/workflows/Dockerfile b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/workflows/Dockerfile
new file mode 100644
index 0000000000..eaffb94f19
--- /dev/null
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/workflows/Dockerfile
@@ -0,0 +1,16 @@
+FROM python:3.12-slim
+
+WORKDIR /app
+
+COPY . user_agent/
+WORKDIR /app/user_agent
+
+RUN if [ -f requirements.txt ]; then \
+        pip install -r requirements.txt; \
+    else \
+        echo "No requirements.txt found"; \
+    fi
+
+EXPOSE 8088
+
+CMD ["python", "main.py"]
\ No newline at end of file
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/workflows/agent.manifest.yaml b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/workflows/agent.manifest.yaml
new file mode 100644
index 0000000000..ad074d6b89
--- /dev/null
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/workflows/agent.manifest.yaml
@@ -0,0 +1,16 @@
+name: echo-agent-responses-streaming
+description: >
+  A simple echo agent that streams responses word-by-word using the
+  azure-ai-agentserver-responses SDK with Server-Sent Events (SSE).
+metadata:
+  tags:
+    - AI Agent Hosting
+    - Azure AI AgentServer
+    - Responses Protocol
+    - Streaming
+template:
+  name: echo-agent-responses-streaming
+  kind: hosted
+  protocols:
+    - protocol: responses
+      version: v0.1.0
\ No newline at end of file
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/workflows/agent.yaml b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/workflows/agent.yaml
new file mode 100644
index 0000000000..3738c690a1
--- /dev/null
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/workflows/agent.yaml
@@ -0,0 +1,8 @@
+kind: hosted
+name: echo-agent-responses-streaming
+protocols:
+  - protocol: responses
+    version: v0.1.0
+resources:
+  cpu: "0.25"
+  memory: 0.5Gi
\ No newline at end of file
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/workflows/main.py b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/workflows/main.py
new file mode 100644
index 0000000000..1b92ec4669
--- /dev/null
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/workflows/main.py
@@ -0,0 +1,49 @@
+# Copyright (c) Microsoft. All rights reserved.
+
+import os
+
+from agent_framework import Agent
+from agent_framework.foundry import FoundryChatClient, ResponsesHostServer
+from azure.ai.agentserver.responses import InMemoryResponseProvider
+from azure.identity import AzureCliCredential
+from dotenv import load_dotenv
+
+# Load environment variables from .env file
+load_dotenv()
+
+
+def main():
+    client = FoundryChatClient(
+        project_endpoint=os.environ["FOUNDRY_PROJECT_ENDPOINT"],
+        model=os.environ["FOUNDRY_MODEL"],
+        credential=AzureCliCredential(),
+    )
+
+    github_pat = os.getenv("GITHUB_PAT")
+    if not github_pat:
+        raise ValueError(
+            "GITHUB_PAT environment variable must be set. Create a token at https://github.com/settings/tokens"
+        )
+
+    github_mcp_tool = client.get_mcp_tool(
+        name="GitHub",
+        url="https://api.githubcopilot.com/mcp/",
+        headers={
+            "Authorization": f"Bearer {github_pat}",
+        },
+        approval_mode="never_require",
+    )
+
+    agent = Agent(
+        client=client,
+        name="HelloAgent",
+        instructions="You are a friendly assistant. Keep your answers brief.",
+        tools=[github_mcp_tool],
+    )
+
+    server = ResponsesHostServer(agent, provider=InMemoryResponseProvider())
+    server.run()
+
+
+if __name__ == "__main__":
+    main()
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/workflows/requirements.txt b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/workflows/requirements.txt
new file mode 100644
index 0000000000..f7dc62f3e3
--- /dev/null
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/workflows/requirements.txt
@@ -0,0 +1,2 @@
+agent-framework
+agent-framework-foundry-hosting
\ No newline at end of file

From b2fa3a63e71414403f61b1b4b0c6ddc975f9d5bb Mon Sep 17 00:00:00 2001
From: Tao Chen <taochen@microsoft.com>
Date: Wed, 8 Apr 2026 14:31:53 -0700
Subject: [PATCH 04/13] Rename samples

---
 .../{helloworld => 01-basic}/.dockerignore       |  0
 .../{helloworld => 01-basic}/Dockerfile          |  0
 .../01-basic/agent.manifest.yaml                 | 15 +++++++++++++++
 .../{hosted_mcp_tools => 01-basic}/agent.yaml    |  2 +-
 .../{helloworld => 01-basic}/main.py             |  5 ++++-
 .../{helloworld => 01-basic}/requirements.txt    |  0
 .../.dockerignore                                |  0
 .../Dockerfile                                   |  0
 .../02_local_tools/agent.manifest.yaml           | 15 +++++++++++++++
 .../{local_tools => 02_local_tools}/agent.yaml   |  2 +-
 .../{local_tools => 02_local_tools}/main.py      |  5 ++++-
 .../requirements.txt                             |  0
 .../{local_tools => 03_remote_mcp}/.dockerignore |  0
 .../{local_tools => 03_remote_mcp}/Dockerfile    |  0
 .../03_remote_mcp/agent.manifest.yaml            | 15 +++++++++++++++
 .../{workflows => 03_remote_mcp}/agent.yaml      |  2 +-
 .../{workflows => 03_remote_mcp}/main.py         |  5 ++++-
 .../requirements.txt                             |  0
 .../{workflows => 04-workflows}/.dockerignore    |  0
 .../{workflows => 04-workflows}/Dockerfile       |  0
 .../04-workflows/agent.manifest.yaml             | 15 +++++++++++++++
 .../{helloworld => 04-workflows}/agent.yaml      |  2 +-
 .../{hosted_mcp_tools => 04-workflows}/main.py   |  8 ++++++++
 .../{workflows => 04-workflows}/requirements.txt |  0
 .../helloworld/agent.manifest.yaml               | 16 ----------------
 .../hosted_mcp_tools/agent.manifest.yaml         | 16 ----------------
 .../local_tools/agent.manifest.yaml              | 16 ----------------
 .../workflows/agent.manifest.yaml                | 16 ----------------
 28 files changed, 84 insertions(+), 71 deletions(-)
 rename python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/{helloworld => 01-basic}/.dockerignore (100%)
 rename python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/{helloworld => 01-basic}/Dockerfile (100%)
 create mode 100644 python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/01-basic/agent.manifest.yaml
 rename python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/{hosted_mcp_tools => 01-basic}/agent.yaml (74%)
 rename python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/{helloworld => 01-basic}/main.py (75%)
 rename python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/{helloworld => 01-basic}/requirements.txt (100%)
 rename python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/{hosted_mcp_tools => 02_local_tools}/.dockerignore (100%)
 rename python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/{hosted_mcp_tools => 02_local_tools}/Dockerfile (100%)
 create mode 100644 python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/02_local_tools/agent.manifest.yaml
 rename python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/{local_tools => 02_local_tools}/agent.yaml (70%)
 rename python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/{local_tools => 02_local_tools}/main.py (82%)
 rename python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/{hosted_mcp_tools => 02_local_tools}/requirements.txt (100%)
 rename python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/{local_tools => 03_remote_mcp}/.dockerignore (100%)
 rename python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/{local_tools => 03_remote_mcp}/Dockerfile (100%)
 create mode 100644 python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/03_remote_mcp/agent.manifest.yaml
 rename python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/{workflows => 03_remote_mcp}/agent.yaml (68%)
 rename python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/{workflows => 03_remote_mcp}/main.py (82%)
 rename python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/{local_tools => 03_remote_mcp}/requirements.txt (100%)
 rename python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/{workflows => 04-workflows}/.dockerignore (100%)
 rename python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/{workflows => 04-workflows}/Dockerfile (100%)
 create mode 100644 python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/04-workflows/agent.manifest.yaml
 rename python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/{helloworld => 04-workflows}/agent.yaml (74%)
 rename python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/{hosted_mcp_tools => 04-workflows}/main.py (79%)
 rename python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/{workflows => 04-workflows}/requirements.txt (100%)
 delete mode 100644 python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/helloworld/agent.manifest.yaml
 delete mode 100644 python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/hosted_mcp_tools/agent.manifest.yaml
 delete mode 100644 python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/local_tools/agent.manifest.yaml
 delete mode 100644 python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/workflows/agent.manifest.yaml

diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/helloworld/.dockerignore b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/01-basic/.dockerignore
similarity index 100%
rename from python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/helloworld/.dockerignore
rename to python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/01-basic/.dockerignore
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/helloworld/Dockerfile b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/01-basic/Dockerfile
similarity index 100%
rename from python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/helloworld/Dockerfile
rename to python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/01-basic/Dockerfile
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/01-basic/agent.manifest.yaml b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/01-basic/agent.manifest.yaml
new file mode 100644
index 0000000000..27bb7630e8
--- /dev/null
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/01-basic/agent.manifest.yaml
@@ -0,0 +1,15 @@
+name: agent-framework-agent-basic
+description: >
+  A basic Agent Framework agent hosted by Foundry.
+metadata:
+  tags:
+    - AI Agent Hosting
+    - Azure AI AgentServer
+    - Responses Protocol
+    - Streaming
+template:
+  name: agent-framework-agent-basic
+  kind: hosted
+  protocols:
+    - protocol: responses
+      version: v0.1.0
\ No newline at end of file
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/hosted_mcp_tools/agent.yaml b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/01-basic/agent.yaml
similarity index 74%
rename from python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/hosted_mcp_tools/agent.yaml
rename to python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/01-basic/agent.yaml
index 3738c690a1..b42f31863a 100644
--- a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/hosted_mcp_tools/agent.yaml
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/01-basic/agent.yaml
@@ -1,5 +1,5 @@
 kind: hosted
-name: echo-agent-responses-streaming
+name: agent-framework-agent-basic
 protocols:
   - protocol: responses
     version: v0.1.0
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/helloworld/main.py b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/01-basic/main.py
similarity index 75%
rename from python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/helloworld/main.py
rename to python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/01-basic/main.py
index 7475b287f5..047719efd9 100644
--- a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/helloworld/main.py
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/01-basic/main.py
@@ -21,8 +21,11 @@ def main():
 
     agent = Agent(
         client=client,
-        name="HelloAgent",
         instructions="You are a friendly assistant. Keep your answers brief.",
+        # History will be managed by the hosting infrastructure, thus there
+        # is no need to store history by the service. Learn more at:
+        # https://developers.openai.com/api/reference/resources/responses/methods/create
+        default_options={"store": False},
     )
 
     server = ResponsesHostServer(agent, provider=InMemoryResponseProvider())
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/helloworld/requirements.txt b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/01-basic/requirements.txt
similarity index 100%
rename from python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/helloworld/requirements.txt
rename to python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/01-basic/requirements.txt
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/hosted_mcp_tools/.dockerignore b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/02_local_tools/.dockerignore
similarity index 100%
rename from python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/hosted_mcp_tools/.dockerignore
rename to python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/02_local_tools/.dockerignore
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/hosted_mcp_tools/Dockerfile b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/02_local_tools/Dockerfile
similarity index 100%
rename from python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/hosted_mcp_tools/Dockerfile
rename to python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/02_local_tools/Dockerfile
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/02_local_tools/agent.manifest.yaml b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/02_local_tools/agent.manifest.yaml
new file mode 100644
index 0000000000..84a6fd95f5
--- /dev/null
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/02_local_tools/agent.manifest.yaml
@@ -0,0 +1,15 @@
+name: agent-framework-agent-with-local-tools
+description: >
+  An Agent Framework agent with local toolshosted by Foundry.
+metadata:
+  tags:
+    - AI Agent Hosting
+    - Azure AI AgentServer
+    - Responses Protocol
+    - Streaming
+template:
+  name: agent-framework-agent-with-local-tools
+  kind: hosted
+  protocols:
+    - protocol: responses
+      version: v0.1.0
\ No newline at end of file
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/local_tools/agent.yaml b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/02_local_tools/agent.yaml
similarity index 70%
rename from python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/local_tools/agent.yaml
rename to python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/02_local_tools/agent.yaml
index 3738c690a1..a2642beb46 100644
--- a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/local_tools/agent.yaml
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/02_local_tools/agent.yaml
@@ -1,5 +1,5 @@
 kind: hosted
-name: echo-agent-responses-streaming
+name: agent-framework-agent-with-local-tools
 protocols:
   - protocol: responses
     version: v0.1.0
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/local_tools/main.py b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/02_local_tools/main.py
similarity index 82%
rename from python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/local_tools/main.py
rename to python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/02_local_tools/main.py
index 690f87d93c..6d213bf8c3 100644
--- a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/local_tools/main.py
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/02_local_tools/main.py
@@ -33,9 +33,12 @@ def main():
 
     agent = Agent(
         client=client,
-        name="HelloAgent",
         instructions="You are a friendly assistant. Keep your answers brief.",
         tools=[get_weather],
+        # History will be managed by the hosting infrastructure, thus there
+        # is no need to store history by the service. Learn more at:
+        # https://developers.openai.com/api/reference/resources/responses/methods/create
+        default_options={"store": False},
     )
 
     server = ResponsesHostServer(agent, provider=InMemoryResponseProvider())
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/hosted_mcp_tools/requirements.txt b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/02_local_tools/requirements.txt
similarity index 100%
rename from python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/hosted_mcp_tools/requirements.txt
rename to python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/02_local_tools/requirements.txt
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/local_tools/.dockerignore b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/03_remote_mcp/.dockerignore
similarity index 100%
rename from python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/local_tools/.dockerignore
rename to python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/03_remote_mcp/.dockerignore
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/local_tools/Dockerfile b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/03_remote_mcp/Dockerfile
similarity index 100%
rename from python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/local_tools/Dockerfile
rename to python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/03_remote_mcp/Dockerfile
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/03_remote_mcp/agent.manifest.yaml b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/03_remote_mcp/agent.manifest.yaml
new file mode 100644
index 0000000000..daf7b10cd3
--- /dev/null
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/03_remote_mcp/agent.manifest.yaml
@@ -0,0 +1,15 @@
+name: agent-framework-agent-with-remote-mcp-tools
+description: >
+  An Agent Framework agent with remote MCP tools hosted by Foundry.
+metadata:
+  tags:
+    - AI Agent Hosting
+    - Azure AI AgentServer
+    - Responses Protocol
+    - Streaming
+template:
+  name: agent-framework-agent-with-remote-mcp-tools
+  kind: hosted
+  protocols:
+    - protocol: responses
+      version: v0.1.0
\ No newline at end of file
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/workflows/agent.yaml b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/03_remote_mcp/agent.yaml
similarity index 68%
rename from python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/workflows/agent.yaml
rename to python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/03_remote_mcp/agent.yaml
index 3738c690a1..9eaedced27 100644
--- a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/workflows/agent.yaml
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/03_remote_mcp/agent.yaml
@@ -1,5 +1,5 @@
 kind: hosted
-name: echo-agent-responses-streaming
+name: agent-framework-agent-with-remote-mcp-tools
 protocols:
   - protocol: responses
     version: v0.1.0
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/workflows/main.py b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/03_remote_mcp/main.py
similarity index 82%
rename from python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/workflows/main.py
rename to python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/03_remote_mcp/main.py
index 1b92ec4669..455b13f1f2 100644
--- a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/workflows/main.py
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/03_remote_mcp/main.py
@@ -36,9 +36,12 @@ def main():
 
     agent = Agent(
         client=client,
-        name="HelloAgent",
         instructions="You are a friendly assistant. Keep your answers brief.",
         tools=[github_mcp_tool],
+        # History will be managed by the hosting infrastructure, thus there
+        # is no need to store history by the service. Learn more at:
+        # https://developers.openai.com/api/reference/resources/responses/methods/create
+        default_options={"store": False},
     )
 
     server = ResponsesHostServer(agent, provider=InMemoryResponseProvider())
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/local_tools/requirements.txt b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/03_remote_mcp/requirements.txt
similarity index 100%
rename from python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/local_tools/requirements.txt
rename to python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/03_remote_mcp/requirements.txt
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/workflows/.dockerignore b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/04-workflows/.dockerignore
similarity index 100%
rename from python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/workflows/.dockerignore
rename to python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/04-workflows/.dockerignore
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/workflows/Dockerfile b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/04-workflows/Dockerfile
similarity index 100%
rename from python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/workflows/Dockerfile
rename to python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/04-workflows/Dockerfile
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/04-workflows/agent.manifest.yaml b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/04-workflows/agent.manifest.yaml
new file mode 100644
index 0000000000..027ca3713b
--- /dev/null
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/04-workflows/agent.manifest.yaml
@@ -0,0 +1,15 @@
+name: agent-framework-workflows
+description: >
+  An Agent Framework workflow hosted by Foundry.
+metadata:
+  tags:
+    - AI Agent Hosting
+    - Azure AI AgentServer
+    - Responses Protocol
+    - Streaming
+template:
+  name: agent-framework-workflows
+  kind: hosted
+  protocols:
+    - protocol: responses
+      version: v0.1.0
\ No newline at end of file
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/helloworld/agent.yaml b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/04-workflows/agent.yaml
similarity index 74%
rename from python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/helloworld/agent.yaml
rename to python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/04-workflows/agent.yaml
index 3738c690a1..ee4fedc2ea 100644
--- a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/helloworld/agent.yaml
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/04-workflows/agent.yaml
@@ -1,5 +1,5 @@
 kind: hosted
-name: echo-agent-responses-streaming
+name: agent-framework-workflows
 protocols:
   - protocol: responses
     version: v0.1.0
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/hosted_mcp_tools/main.py b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/04-workflows/main.py
similarity index 79%
rename from python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/hosted_mcp_tools/main.py
rename to python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/04-workflows/main.py
index 164ef87fcf..9d6ed68158 100644
--- a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/hosted_mcp_tools/main.py
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/04-workflows/main.py
@@ -33,6 +33,10 @@ def main():
             "You are an excellent content writer. You create new content and edit contents based on the feedback."
         ),
         name="writer",
+        # History will be managed by the hosting infrastructure, thus there
+        # is no need to store history by the service. Learn more at:
+        # https://developers.openai.com/api/reference/resources/responses/methods/create
+        default_options={"store": False},
     )
 
     reviewer_agent = Agent(
@@ -43,6 +47,10 @@ def main():
             "Provide the feedback in the most concise manner possible."
         ),
         name="reviewer",
+        # History will be managed by the hosting infrastructure, thus there
+        # is no need to store history by the service. Learn more at:
+        # https://developers.openai.com/api/reference/resources/responses/methods/create
+        default_options={"store": False},
     )
 
     workflow_agent = (
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/workflows/requirements.txt b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/04-workflows/requirements.txt
similarity index 100%
rename from python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/workflows/requirements.txt
rename to python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/04-workflows/requirements.txt
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/helloworld/agent.manifest.yaml b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/helloworld/agent.manifest.yaml
deleted file mode 100644
index ad074d6b89..0000000000
--- a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/helloworld/agent.manifest.yaml
+++ /dev/null
@@ -1,16 +0,0 @@
-name: echo-agent-responses-streaming
-description: >
-  A simple echo agent that streams responses word-by-word using the
-  azure-ai-agentserver-responses SDK with Server-Sent Events (SSE).
-metadata:
-  tags:
-    - AI Agent Hosting
-    - Azure AI AgentServer
-    - Responses Protocol
-    - Streaming
-template:
-  name: echo-agent-responses-streaming
-  kind: hosted
-  protocols:
-    - protocol: responses
-      version: v0.1.0
\ No newline at end of file
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/hosted_mcp_tools/agent.manifest.yaml b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/hosted_mcp_tools/agent.manifest.yaml
deleted file mode 100644
index ad074d6b89..0000000000
--- a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/hosted_mcp_tools/agent.manifest.yaml
+++ /dev/null
@@ -1,16 +0,0 @@
-name: echo-agent-responses-streaming
-description: >
-  A simple echo agent that streams responses word-by-word using the
-  azure-ai-agentserver-responses SDK with Server-Sent Events (SSE).
-metadata:
-  tags:
-    - AI Agent Hosting
-    - Azure AI AgentServer
-    - Responses Protocol
-    - Streaming
-template:
-  name: echo-agent-responses-streaming
-  kind: hosted
-  protocols:
-    - protocol: responses
-      version: v0.1.0
\ No newline at end of file
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/local_tools/agent.manifest.yaml b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/local_tools/agent.manifest.yaml
deleted file mode 100644
index ad074d6b89..0000000000
--- a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/local_tools/agent.manifest.yaml
+++ /dev/null
@@ -1,16 +0,0 @@
-name: echo-agent-responses-streaming
-description: >
-  A simple echo agent that streams responses word-by-word using the
-  azure-ai-agentserver-responses SDK with Server-Sent Events (SSE).
-metadata:
-  tags:
-    - AI Agent Hosting
-    - Azure AI AgentServer
-    - Responses Protocol
-    - Streaming
-template:
-  name: echo-agent-responses-streaming
-  kind: hosted
-  protocols:
-    - protocol: responses
-      version: v0.1.0
\ No newline at end of file
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/workflows/agent.manifest.yaml b/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/workflows/agent.manifest.yaml
deleted file mode 100644
index ad074d6b89..0000000000
--- a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/workflows/agent.manifest.yaml
+++ /dev/null
@@ -1,16 +0,0 @@
-name: echo-agent-responses-streaming
-description: >
-  A simple echo agent that streams responses word-by-word using the
-  azure-ai-agentserver-responses SDK with Server-Sent Events (SSE).
-metadata:
-  tags:
-    - AI Agent Hosting
-    - Azure AI AgentServer
-    - Responses Protocol
-    - Streaming
-template:
-  name: echo-agent-responses-streaming
-  kind: hosted
-  protocols:
-    - protocol: responses
-      version: v0.1.0
\ No newline at end of file

From 8403d57cafe363969e7339356f6e25520f3c5e01 Mon Sep 17 00:00:00 2001
From: Tao Chen <taochen@microsoft.com>
Date: Wed, 8 Apr 2026 16:48:24 -0700
Subject: [PATCH 05/13] Add invocations

---
 .../core/agent_framework/foundry/__init__.py  |  1 +
 .../core/agent_framework/foundry/__init__.pyi |  3 +-
 .../__init__.py                               |  3 +-
 .../_invocations.py                           | 73 +++++++++++++++++++
 .../_responses.py                             | 14 ++--
 .../01-basic/README.md                        | 13 ++++
 .../01-basic/main.py                          | 35 +++++++++
 .../01-basic/requirements.txt                 |  0
 .../01-basic/.dockerignore                    |  0
 .../01-basic/Dockerfile                       |  0
 .../01-basic/README.md                        | 19 +++++
 .../01-basic/agent.manifest.yaml              |  0
 .../01-basic/agent.yaml                       |  0
 .../01-basic/main.py                          |  0
 .../01-basic}/requirements.txt                |  0
 .../02_local_tools/.dockerignore              |  0
 .../02_local_tools/Dockerfile                 |  0
 .../02_local_tools/README.md                  | 13 ++++
 .../02_local_tools/agent.manifest.yaml        |  0
 .../02_local_tools/agent.yaml                 |  0
 .../02_local_tools/main.py                    |  0
 .../02_local_tools}/requirements.txt          |  0
 .../03_remote_mcp/.dockerignore               |  0
 .../03_remote_mcp/Dockerfile                  |  0
 .../03_remote_mcp/README.md                   | 13 ++++
 .../03_remote_mcp/agent.manifest.yaml         |  0
 .../03_remote_mcp/agent.yaml                  |  0
 .../03_remote_mcp/main.py                     |  0
 .../03_remote_mcp}/requirements.txt           |  0
 .../04-workflows/.dockerignore                |  0
 .../04-workflows/Dockerfile                   |  0
 .../04-workflows/README.md                    | 13 ++++
 .../04-workflows/agent.manifest.yaml          |  0
 .../04-workflows/agent.yaml                   |  0
 .../04-workflows/main.py                      |  0
 .../04-workflows/requirements.txt             |  2 +
 36 files changed, 194 insertions(+), 8 deletions(-)
 create mode 100644 python/packages/foundry_hosting/agent_framework_foundry_hosting/_invocations.py
 create mode 100644 python/samples/05-end-to-end/hosted_agents/foundry_hosted_invocations/01-basic/README.md
 create mode 100644 python/samples/05-end-to-end/hosted_agents/foundry_hosted_invocations/01-basic/main.py
 rename python/samples/05-end-to-end/hosted_agents/{foundry_vnext_responses => foundry_hosted_invocations}/01-basic/requirements.txt (100%)
 rename python/samples/05-end-to-end/hosted_agents/{foundry_vnext_responses => foundry_hosted_responses}/01-basic/.dockerignore (100%)
 rename python/samples/05-end-to-end/hosted_agents/{foundry_vnext_responses => foundry_hosted_responses}/01-basic/Dockerfile (100%)
 create mode 100644 python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/01-basic/README.md
 rename python/samples/05-end-to-end/hosted_agents/{foundry_vnext_responses => foundry_hosted_responses}/01-basic/agent.manifest.yaml (100%)
 rename python/samples/05-end-to-end/hosted_agents/{foundry_vnext_responses => foundry_hosted_responses}/01-basic/agent.yaml (100%)
 rename python/samples/05-end-to-end/hosted_agents/{foundry_vnext_responses => foundry_hosted_responses}/01-basic/main.py (100%)
 rename python/samples/05-end-to-end/hosted_agents/{foundry_vnext_responses/02_local_tools => foundry_hosted_responses/01-basic}/requirements.txt (100%)
 rename python/samples/05-end-to-end/hosted_agents/{foundry_vnext_responses => foundry_hosted_responses}/02_local_tools/.dockerignore (100%)
 rename python/samples/05-end-to-end/hosted_agents/{foundry_vnext_responses => foundry_hosted_responses}/02_local_tools/Dockerfile (100%)
 create mode 100644 python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/02_local_tools/README.md
 rename python/samples/05-end-to-end/hosted_agents/{foundry_vnext_responses => foundry_hosted_responses}/02_local_tools/agent.manifest.yaml (100%)
 rename python/samples/05-end-to-end/hosted_agents/{foundry_vnext_responses => foundry_hosted_responses}/02_local_tools/agent.yaml (100%)
 rename python/samples/05-end-to-end/hosted_agents/{foundry_vnext_responses => foundry_hosted_responses}/02_local_tools/main.py (100%)
 rename python/samples/05-end-to-end/hosted_agents/{foundry_vnext_responses/03_remote_mcp => foundry_hosted_responses/02_local_tools}/requirements.txt (100%)
 rename python/samples/05-end-to-end/hosted_agents/{foundry_vnext_responses => foundry_hosted_responses}/03_remote_mcp/.dockerignore (100%)
 rename python/samples/05-end-to-end/hosted_agents/{foundry_vnext_responses => foundry_hosted_responses}/03_remote_mcp/Dockerfile (100%)
 create mode 100644 python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/03_remote_mcp/README.md
 rename python/samples/05-end-to-end/hosted_agents/{foundry_vnext_responses => foundry_hosted_responses}/03_remote_mcp/agent.manifest.yaml (100%)
 rename python/samples/05-end-to-end/hosted_agents/{foundry_vnext_responses => foundry_hosted_responses}/03_remote_mcp/agent.yaml (100%)
 rename python/samples/05-end-to-end/hosted_agents/{foundry_vnext_responses => foundry_hosted_responses}/03_remote_mcp/main.py (100%)
 rename python/samples/05-end-to-end/hosted_agents/{foundry_vnext_responses/04-workflows => foundry_hosted_responses/03_remote_mcp}/requirements.txt (100%)
 rename python/samples/05-end-to-end/hosted_agents/{foundry_vnext_responses => foundry_hosted_responses}/04-workflows/.dockerignore (100%)
 rename python/samples/05-end-to-end/hosted_agents/{foundry_vnext_responses => foundry_hosted_responses}/04-workflows/Dockerfile (100%)
 create mode 100644 python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/04-workflows/README.md
 rename python/samples/05-end-to-end/hosted_agents/{foundry_vnext_responses => foundry_hosted_responses}/04-workflows/agent.manifest.yaml (100%)
 rename python/samples/05-end-to-end/hosted_agents/{foundry_vnext_responses => foundry_hosted_responses}/04-workflows/agent.yaml (100%)
 rename python/samples/05-end-to-end/hosted_agents/{foundry_vnext_responses => foundry_hosted_responses}/04-workflows/main.py (100%)
 create mode 100644 python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/04-workflows/requirements.txt

diff --git a/python/packages/core/agent_framework/foundry/__init__.py b/python/packages/core/agent_framework/foundry/__init__.py
index f438b2ab10..eb9231f1da 100644
--- a/python/packages/core/agent_framework/foundry/__init__.py
+++ b/python/packages/core/agent_framework/foundry/__init__.py
@@ -33,6 +33,7 @@
     "evaluate_foundry_target": ("agent_framework_foundry", "agent-framework-foundry"),
     "evaluate_traces": ("agent_framework_foundry", "agent-framework-foundry"),
     "ResponsesHostServer": ("agent_framework_foundry_hosting", "agent-framework-foundry-hosting"),
+    "InvocationsHostServer": ("agent_framework_foundry_hosting", "agent-framework-foundry-hosting"),
 }
 
 
diff --git a/python/packages/core/agent_framework/foundry/__init__.pyi b/python/packages/core/agent_framework/foundry/__init__.pyi
index 2d2a5e2920..3e7512b3aa 100644
--- a/python/packages/core/agent_framework/foundry/__init__.pyi
+++ b/python/packages/core/agent_framework/foundry/__init__.pyi
@@ -20,7 +20,7 @@ from agent_framework_foundry import (
     evaluate_foundry_target,
     evaluate_traces,
 )
-from agent_framework_foundry_hosting import ResponsesHostServer
+from agent_framework_foundry_hosting import InvocationsHostServer, ResponsesHostServer
 from agent_framework_foundry_local import (
     FoundryLocalChatOptions,
     FoundryLocalClient,
@@ -40,6 +40,7 @@ __all__ = [
     "FoundryLocalClient",
     "FoundryLocalSettings",
     "FoundryMemoryProvider",
+    "InvocationsHostServer",
     "RawAnthropicFoundryClient",
     "RawFoundryAgent",
     "RawFoundryAgentChatClient",
diff --git a/python/packages/foundry_hosting/agent_framework_foundry_hosting/__init__.py b/python/packages/foundry_hosting/agent_framework_foundry_hosting/__init__.py
index bad5e56fe0..81e8430783 100644
--- a/python/packages/foundry_hosting/agent_framework_foundry_hosting/__init__.py
+++ b/python/packages/foundry_hosting/agent_framework_foundry_hosting/__init__.py
@@ -2,6 +2,7 @@
 
 import importlib.metadata
 
+from ._invocations import InvocationsHostServer
 from ._responses import ResponsesHostServer
 
 try:
@@ -9,4 +10,4 @@
 except importlib.metadata.PackageNotFoundError:
     __version__ = "0.0.0"
 
-__all__ = ["ResponsesHostServer"]
+__all__ = ["InvocationsHostServer", "ResponsesHostServer"]
diff --git a/python/packages/foundry_hosting/agent_framework_foundry_hosting/_invocations.py b/python/packages/foundry_hosting/agent_framework_foundry_hosting/_invocations.py
new file mode 100644
index 0000000000..ae3330fff4
--- /dev/null
+++ b/python/packages/foundry_hosting/agent_framework_foundry_hosting/_invocations.py
@@ -0,0 +1,73 @@
+# Copyright (c) Microsoft. All rights reserved.
+
+from agent_framework import AgentSession, BaseAgent, SupportsAgentRun
+from azure.ai.agentserver.invocations import InvocationAgentServerHost
+from starlette.requests import Request
+from starlette.responses import JSONResponse, Response, StreamingResponse
+from typing_extensions import Any, AsyncGenerator, Optional
+
+
+class InvocationsHostServer(InvocationAgentServerHost):
+    """An invocations server host for an agent."""
+
+    def __init__(
+        self,
+        agent: BaseAgent,
+        *,
+        stream: bool = False,
+        openapi_spec: Optional[dict[str, Any]] = None,
+        **kwargs: Any,
+    ) -> None:
+        """Initialize an InvocationsHostServer.
+
+        Args:
+            agent: The agent to handle responses for.
+            stream: Whether to stream the responses. Defaults to True.
+            openapi_spec: The OpenAPI specification for the server.
+            **kwargs: Additional keyword arguments.
+
+        This host will expect the request to be a JSON body with a "message" field.
+        The response from the host will be a JSON object with a "response" field containing
+        the agent's response and a "session_id" field containing the session ID.
+        """
+        super().__init__(openapi_spec=openapi_spec, **kwargs)
+
+        if not isinstance(agent, SupportsAgentRun):
+            raise TypeError("Agent must support the SupportsAgentRun interface")
+
+        self._agent = agent
+        self._stream = stream
+        self._sessions: dict[str, AgentSession] = {}
+        self.invoke_handler(self._handle_invoke)  # pyright: ignore[reportUnknownMemberType]
+
+    async def _handle_invoke(self, request: Request) -> Response:
+        """Invoke the agent with the given request."""
+        data = await request.json()
+        session_id: str = request.state.session_id
+
+        user_message = data.get("message", None)
+        if user_message is None:
+            error = "Missing 'message' in request"
+            if self._stream:
+                return StreamingResponse(content=error, status_code=400)
+            return Response(content=error, status_code=400)
+
+        session = self._sessions.setdefault(session_id, AgentSession(session_id=session_id))
+
+        if self._stream:
+
+            async def stream_response() -> AsyncGenerator[str]:
+                async for update in self._agent.run(user_message, session=session, stream=True):
+                    yield update.text
+
+            return StreamingResponse(
+                stream_response(),
+                media_type="text/event-stream",
+                headers={"Cache-Control": "no-cache", "Connection": "keep-alive"},
+            )
+
+        response = await self._agent.run([user_message], session=session, stream=self._stream)
+        return JSONResponse({
+            "response": response.text,
+            "session_id": session_id,
+        })
diff --git a/python/packages/foundry_hosting/agent_framework_foundry_hosting/_responses.py b/python/packages/foundry_hosting/agent_framework_foundry_hosting/_responses.py
index eca938afbd..b865b21d9a 100644
--- a/python/packages/foundry_hosting/agent_framework_foundry_hosting/_responses.py
+++ b/python/packages/foundry_hosting/agent_framework_foundry_hosting/_responses.py
@@ -81,7 +81,7 @@ def __init__(
         if not isinstance(agent, SupportsAgentRun):
             raise TypeError("Agent must support the SupportsAgentRun interface")
 
-        self.agent = agent
+        self._agent = agent
         self.create_handler(self._handle_create)  # pyright: ignore[reportUnknownMemberType]
 
     async def _handle_create(
@@ -92,17 +92,17 @@ async def _handle_create(
     ) -> AsyncIterable[dict[str, Any]]:
         # Replace or add a history provider that has `load_messages=True`
         history_provider_idx: list[int] = []
-        for i, provider in enumerate(self.agent.context_providers):
+        for i, provider in enumerate(self._agent.context_providers):
             if isinstance(provider, HistoryProvider) and provider.load_messages:
                 history_provider_idx.append(i)
 
         if not history_provider_idx:
-            self.agent.context_providers.append(ResponsesHostContextProvider(context))
+            self._agent.context_providers.append(ResponsesHostContextProvider(context))
         elif len(history_provider_idx) > 1:
             # There shouldn't be more than one history provider with `load_messages=True`
             raise RuntimeError("There shouldn't be more than one history provider with `load_messages=True`")
         else:
-            self.agent.context_providers[history_provider_idx[0]] = ResponsesHostContextProvider(context)
+            self._agent.context_providers[history_provider_idx[0]] = ResponsesHostContextProvider(context)
 
         input_items = get_input_text(request)
 
@@ -111,9 +111,11 @@ async def _handle_create(
         yield stream.emit_created()
         yield stream.emit_in_progress()
 
+        # Add reasoning
+
         if request.stream is None or request.stream is False:
             # Run the agent in non-streaming mode
-            response = await self.agent.run(input_items, stream=False)
+            response = await self._agent.run(input_items, stream=False)
             for item in stream.output_item_message(response.text):
                 yield item
             yield stream.emit_completed()
@@ -126,7 +128,7 @@ async def _handle_create(
 
         # Invoke the MAF agent
         full_text = ""
-        async for update in self.agent.run(input_items, stream=True):
+        async for update in self._agent.run(input_items, stream=True):
             full_text += update.text
             yield text_content.emit_delta(update.text)
 
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_hosted_invocations/01-basic/README.md b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_invocations/01-basic/README.md
new file mode 100644
index 0000000000..040f562f64
--- /dev/null
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_invocations/01-basic/README.md
@@ -0,0 +1,13 @@
+# Basic example of hosting an agent with the `invocations` API
+
+Run the following command to start the server:
+
+```bash
+python main.py
+```
+
+Send a POST request to the server with a JSON body containing a "message" field to interact with the agent. For example:
+
+```bash
+curl -X POST http://localhost:8088/invocations -H "Content-Type: application/json" -d '{"message": "Hi!"}'
+```
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_hosted_invocations/01-basic/main.py b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_invocations/01-basic/main.py
new file mode 100644
index 0000000000..4145aee14e
--- /dev/null
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_invocations/01-basic/main.py
@@ -0,0 +1,35 @@
+# Copyright (c) Microsoft. All rights reserved.
+
+import os
+
+from agent_framework import Agent
+from agent_framework.foundry import FoundryChatClient, InvocationsHostServer
+from azure.identity import AzureCliCredential
+from dotenv import load_dotenv
+
+# Load environment variables from .env file
+load_dotenv()
+
+
+def main():
+    client = FoundryChatClient(
+        project_endpoint=os.environ["FOUNDRY_PROJECT_ENDPOINT"],
+        model=os.environ["FOUNDRY_MODEL"],
+        credential=AzureCliCredential(),
+    )
+
+    agent = Agent(
+        client=client,
+        instructions="You are a friendly assistant. Keep your answers brief.",
+        # History will be managed by the hosting infrastructure, thus there
+        # is no need to store history by the service. Learn more at:
+        # https://developers.openai.com/api/reference/resources/responses/methods/create
+        default_options={"store": False},
+    )
+
+    server = InvocationsHostServer(agent)
+    server.run()
+
+
+if __name__ == "__main__":
+    main()
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/01-basic/requirements.txt b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_invocations/01-basic/requirements.txt
similarity index 100%
rename from python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/01-basic/requirements.txt
rename to python/samples/05-end-to-end/hosted_agents/foundry_hosted_invocations/01-basic/requirements.txt
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/01-basic/.dockerignore b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/01-basic/.dockerignore
similarity index 100%
rename from python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/01-basic/.dockerignore
rename to python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/01-basic/.dockerignore
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/01-basic/Dockerfile b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/01-basic/Dockerfile
similarity index 100%
rename from python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/01-basic/Dockerfile
rename to python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/01-basic/Dockerfile
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/01-basic/README.md b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/01-basic/README.md
new file mode 100644
index 0000000000..57686cddda
--- /dev/null
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/01-basic/README.md
@@ -0,0 +1,19 @@
+# Basic example of hosting an agent with the `responses` API
+
+Run the following command to start the server:
+
+```bash
+python main.py
+```
+
+Send a POST request to the server with a JSON body containing a "message" field to interact with the agent. For example:
+
+```bash
+curl -X POST http://localhost:8088/responses -H "Content-Type: application/json" -d '{"input": "Hi"}'
+```
+
+To have a multi-turn conversation with the agent, include the previous response id in the request body. For example:
+
+```bash
+curl -X POST http://localhost:8088/responses -H "Content-Type: application/json" -d '{"input": "How are you?", "previous_response_id": "REPLACE_WITH_PREVIOUS_RESPONSE_ID"}'
+```
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/01-basic/agent.manifest.yaml b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/01-basic/agent.manifest.yaml
similarity index 100%
rename from python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/01-basic/agent.manifest.yaml
rename to python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/01-basic/agent.manifest.yaml
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/01-basic/agent.yaml b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/01-basic/agent.yaml
similarity index 100%
rename from python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/01-basic/agent.yaml
rename to python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/01-basic/agent.yaml
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/01-basic/main.py b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/01-basic/main.py
similarity index 100%
rename from python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/01-basic/main.py
rename to python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/01-basic/main.py
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/02_local_tools/requirements.txt b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/01-basic/requirements.txt
similarity index 100%
rename from python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/02_local_tools/requirements.txt
rename to python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/01-basic/requirements.txt
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/02_local_tools/.dockerignore b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/02_local_tools/.dockerignore
similarity index 100%
rename from python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/02_local_tools/.dockerignore
rename to python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/02_local_tools/.dockerignore
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/02_local_tools/Dockerfile b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/02_local_tools/Dockerfile
similarity index 100%
rename from python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/02_local_tools/Dockerfile
rename to python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/02_local_tools/Dockerfile
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/02_local_tools/README.md b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/02_local_tools/README.md
new file mode 100644
index 0000000000..b262afa7ab
--- /dev/null
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/02_local_tools/README.md
@@ -0,0 +1,13 @@
+# Basic example of hosting an agent with the `responses` API and local tools
+
+Run the following command to start the server:
+
+```bash
+python main.py
+```
+
+Send a POST request to the server with a JSON body containing a "message" field to interact with the agent. For example:
+
+```bash
+curl -X POST http://localhost:8088/responses -H "Content-Type: application/json" -d '{"input": "What is the weather in Seattle?"}'
+```
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/02_local_tools/agent.manifest.yaml b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/02_local_tools/agent.manifest.yaml
similarity index 100%
rename from python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/02_local_tools/agent.manifest.yaml
rename to python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/02_local_tools/agent.manifest.yaml
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/02_local_tools/agent.yaml b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/02_local_tools/agent.yaml
similarity index 100%
rename from python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/02_local_tools/agent.yaml
rename to python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/02_local_tools/agent.yaml
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/02_local_tools/main.py b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/02_local_tools/main.py
similarity index 100%
rename from python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/02_local_tools/main.py
rename to python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/02_local_tools/main.py
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/03_remote_mcp/requirements.txt b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/02_local_tools/requirements.txt
similarity index 100%
rename from python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/03_remote_mcp/requirements.txt
rename to python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/02_local_tools/requirements.txt
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/03_remote_mcp/.dockerignore b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/03_remote_mcp/.dockerignore
similarity index 100%
rename from python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/03_remote_mcp/.dockerignore
rename to python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/03_remote_mcp/.dockerignore
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/03_remote_mcp/Dockerfile b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/03_remote_mcp/Dockerfile
similarity index 100%
rename from python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/03_remote_mcp/Dockerfile
rename to python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/03_remote_mcp/Dockerfile
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/03_remote_mcp/README.md b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/03_remote_mcp/README.md
new file mode 100644
index 0000000000..5091871aa8
--- /dev/null
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/03_remote_mcp/README.md
@@ -0,0 +1,13 @@
+# Basic example of hosting an agent with the `responses` API and a remote MCP
+
+Run the following command to start the server:
+
+```bash
+python main.py
+```
+
+Send a POST request to the server with a JSON body containing a "message" field to interact with the agent. For example:
+
+```bash
+curl -X POST http://localhost:8088/responses -H "Content-Type: application/json" -d '{"input": "List all the repositories I own on GitHub."}'
+```
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/03_remote_mcp/agent.manifest.yaml b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/03_remote_mcp/agent.manifest.yaml
similarity index 100%
rename from python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/03_remote_mcp/agent.manifest.yaml
rename to python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/03_remote_mcp/agent.manifest.yaml
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/03_remote_mcp/agent.yaml b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/03_remote_mcp/agent.yaml
similarity index 100%
rename from python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/03_remote_mcp/agent.yaml
rename to python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/03_remote_mcp/agent.yaml
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/03_remote_mcp/main.py b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/03_remote_mcp/main.py
similarity index 100%
rename from python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/03_remote_mcp/main.py
rename to python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/03_remote_mcp/main.py
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/04-workflows/requirements.txt b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/03_remote_mcp/requirements.txt
similarity index 100%
rename from python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/04-workflows/requirements.txt
rename to python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/03_remote_mcp/requirements.txt
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/04-workflows/.dockerignore b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/04-workflows/.dockerignore
similarity index 100%
rename from python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/04-workflows/.dockerignore
rename to python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/04-workflows/.dockerignore
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/04-workflows/Dockerfile b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/04-workflows/Dockerfile
similarity index 100%
rename from python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/04-workflows/Dockerfile
rename to python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/04-workflows/Dockerfile
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/04-workflows/README.md b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/04-workflows/README.md
new file mode 100644
index 0000000000..75d87c3fcb
--- /dev/null
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/04-workflows/README.md
@@ -0,0 +1,13 @@
+# Basic example of hosting an agent with the `responses` API and a workflow
+
+Run the following command to start the server:
+
+```bash
+python main.py
+```
+
+Send a POST request to the server with a JSON body containing a "message" field to interact with the agent. For example:
+
+```bash
+curl -X POST http://localhost:8088/responses -H "Content-Type: application/json" -d '{"input": "Create a slogan for a new electric SUV that is affordable and fun to drive."}'
+```
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/04-workflows/agent.manifest.yaml b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/04-workflows/agent.manifest.yaml
similarity index 100%
rename from python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/04-workflows/agent.manifest.yaml
rename to python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/04-workflows/agent.manifest.yaml
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/04-workflows/agent.yaml b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/04-workflows/agent.yaml
similarity index 100%
rename from python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/04-workflows/agent.yaml
rename to python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/04-workflows/agent.yaml
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/04-workflows/main.py b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/04-workflows/main.py
similarity index 100%
rename from python/samples/05-end-to-end/hosted_agents/foundry_vnext_responses/04-workflows/main.py
rename to python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/04-workflows/main.py
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/04-workflows/requirements.txt b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/04-workflows/requirements.txt
new file mode 100644
index 0000000000..f7dc62f3e3
--- /dev/null
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/04-workflows/requirements.txt
@@ -0,0 +1,2 @@
+agent-framework
+agent-framework-foundry-hosting
\ No newline at end of file

From ea625a7f5aca36461e1b2c27097ad28789435545 Mon Sep 17 00:00:00 2001
From: Tao Chen <taochen@microsoft.com>
Date: Thu, 9 Apr 2026 09:04:34 -0700
Subject: [PATCH 06/13] Comments 1

---
 .../core/agent_framework/foundry/__init__.py         |  3 ---
 .../core/agent_framework/foundry/__init__.pyi        |  3 ---
 python/packages/core/pyproject.toml                  |  1 -
 .../{_shared.py => _message_converter.py}            |  0
 .../agent_framework_foundry_hosting/_responses.py    | 12 ++++++------
 python/packages/foundry_hosting/pyproject.toml       |  4 ++--
 .../foundry_hosted_invocations/01-basic/main.py      |  3 ++-
 .../foundry_hosted_responses/01-basic/main.py        |  3 ++-
 .../foundry_hosted_responses/02_local_tools/main.py  |  3 ++-
 .../foundry_hosted_responses/03_remote_mcp/main.py   |  3 ++-
 .../foundry_hosted_responses/04-workflows/main.py    |  3 ++-
 11 files changed, 18 insertions(+), 20 deletions(-)
 rename python/packages/foundry_hosting/agent_framework_foundry_hosting/{_shared.py => _message_converter.py} (100%)

diff --git a/python/packages/core/agent_framework/foundry/__init__.py b/python/packages/core/agent_framework/foundry/__init__.py
index eb9231f1da..b1d2b88450 100644
--- a/python/packages/core/agent_framework/foundry/__init__.py
+++ b/python/packages/core/agent_framework/foundry/__init__.py
@@ -6,7 +6,6 @@
 - ``agent-framework-anthropic``
 - ``agent-framework-foundry``
 - ``agent-framework-foundry-local``
-- ``agent-framework-foundry-hosting``
 """
 
 import importlib
@@ -32,8 +31,6 @@
     "RawFoundryEmbeddingClient": ("agent_framework_foundry", "agent-framework-foundry"),
     "evaluate_foundry_target": ("agent_framework_foundry", "agent-framework-foundry"),
     "evaluate_traces": ("agent_framework_foundry", "agent-framework-foundry"),
-    "ResponsesHostServer": ("agent_framework_foundry_hosting", "agent-framework-foundry-hosting"),
-    "InvocationsHostServer": ("agent_framework_foundry_hosting", "agent-framework-foundry-hosting"),
 }
 
 
diff --git a/python/packages/core/agent_framework/foundry/__init__.pyi b/python/packages/core/agent_framework/foundry/__init__.pyi
index 3e7512b3aa..47eb92b3af 100644
--- a/python/packages/core/agent_framework/foundry/__init__.pyi
+++ b/python/packages/core/agent_framework/foundry/__init__.pyi
@@ -20,7 +20,6 @@ from agent_framework_foundry import (
     evaluate_foundry_target,
     evaluate_traces,
 )
-from agent_framework_foundry_hosting import InvocationsHostServer, ResponsesHostServer
 from agent_framework_foundry_local import (
     FoundryLocalChatOptions,
     FoundryLocalClient,
@@ -40,13 +39,11 @@ __all__ = [
     "FoundryLocalClient",
     "FoundryLocalSettings",
     "FoundryMemoryProvider",
-    "InvocationsHostServer",
     "RawAnthropicFoundryClient",
     "RawFoundryAgent",
     "RawFoundryAgentChatClient",
     "RawFoundryChatClient",
     "RawFoundryEmbeddingClient",
-    "ResponsesHostServer",
     "evaluate_foundry_target",
     "evaluate_traces",
 ]
diff --git a/python/packages/core/pyproject.toml b/python/packages/core/pyproject.toml
index c6e7ef2b42..e4aa1f5e40 100644
--- a/python/packages/core/pyproject.toml
+++ b/python/packages/core/pyproject.toml
@@ -48,7 +48,6 @@ all = [
     "agent-framework-durabletask",
     "agent-framework-foundry",
     "agent-framework-foundry-local",
-    "agent-framework-foundry-hosting",
     "agent-framework-github-copilot; python_version >= '3.11'",
     "agent-framework-lab",
     "agent-framework-mem0",
diff --git a/python/packages/foundry_hosting/agent_framework_foundry_hosting/_shared.py b/python/packages/foundry_hosting/agent_framework_foundry_hosting/_message_converter.py
similarity index 100%
rename from python/packages/foundry_hosting/agent_framework_foundry_hosting/_shared.py
rename to python/packages/foundry_hosting/agent_framework_foundry_hosting/_message_converter.py
diff --git a/python/packages/foundry_hosting/agent_framework_foundry_hosting/_responses.py b/python/packages/foundry_hosting/agent_framework_foundry_hosting/_responses.py
index b865b21d9a..221ae279ca 100644
--- a/python/packages/foundry_hosting/agent_framework_foundry_hosting/_responses.py
+++ b/python/packages/foundry_hosting/agent_framework_foundry_hosting/_responses.py
@@ -14,14 +14,14 @@
 from azure.ai.agentserver.responses.models import CreateResponse, get_input_text
 from typing_extensions import Any, Sequence
 
-from ._shared import to_messages
+from ._message_converter import to_messages
 
 
-class ResponsesHostContextProvider(HistoryProvider):
+class ResponsesHostHistoryProvider(HistoryProvider):
     """A history provider that retrieves messages from a ResponseContext."""
 
     def __init__(self, context: ResponseContext):
-        """Initialize a ResponsesHostContextProvider.
+        """Initialize a ResponsesHostHistoryProvider.
 
         Args:
             context: The ResponseContext to retrieve messages from.
@@ -73,7 +73,7 @@ def __init__(
 
         Note:
             If the agent has a history provider with `load_messages=True`, it will be
-            replaced with a `ResponsesHostContextProvider` that will retrieve history
+            replaced with a `ResponsesHostHistoryProvider` that will retrieve history
             from the hosting infrastructure.
         """
         super().__init__(prefix=prefix, options=options, provider=provider, **kwargs)
@@ -97,12 +97,12 @@ async def _handle_create(
                 history_provider_idx.append(i)
 
         if not history_provider_idx:
-            self._agent.context_providers.append(ResponsesHostContextProvider(context))
+            self._agent.context_providers.append(ResponsesHostHistoryProvider(context))
         elif len(history_provider_idx) > 1:
             # There shouldn't be more than one history provider with `load_messages=True`
             raise RuntimeError("There shouldn't be more than one history provider with `load_messages=True`")
         else:
-            self._agent.context_providers[history_provider_idx[0]] = ResponsesHostContextProvider(context)
+            self._agent.context_providers[history_provider_idx[0]] = ResponsesHostHistoryProvider(context)
 
         input_items = get_input_text(request)
 
diff --git a/python/packages/foundry_hosting/pyproject.toml b/python/packages/foundry_hosting/pyproject.toml
index 1f0c42dbbf..3bb6c71495 100644
--- a/python/packages/foundry_hosting/pyproject.toml
+++ b/python/packages/foundry_hosting/pyproject.toml
@@ -1,6 +1,6 @@
 [project]
 name = "agent-framework-foundry-hosting"
-description = "Foundry Local integration for Microsoft Agent Framework."
+description = "Foundry Hosting integration for Microsoft Agent Framework."
 authors = [{ name = "Microsoft", email = "af-support@microsoft.com"}]
 readme = "README.md"
 requires-python = ">=3.10"
@@ -12,7 +12,7 @@ urls.release_notes = "https://github.com/microsoft/agent-framework/releases?q=ta
 urls.issues = "https://github.com/microsoft/agent-framework/issues"
 classifiers = [
   "License :: OSI Approved :: MIT License",
-  "Development Status :: 4 - Beta",
+  "Development Status :: 4 - Alpha",
   "Intended Audience :: Developers",
   "Programming Language :: Python :: 3",
   "Programming Language :: Python :: 3.10",
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_hosted_invocations/01-basic/main.py b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_invocations/01-basic/main.py
index 4145aee14e..17f9cb6341 100644
--- a/python/samples/05-end-to-end/hosted_agents/foundry_hosted_invocations/01-basic/main.py
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_invocations/01-basic/main.py
@@ -3,7 +3,8 @@
 import os
 
 from agent_framework import Agent
-from agent_framework.foundry import FoundryChatClient, InvocationsHostServer
+from agent_framework.foundry import FoundryChatClient
+from agent_framework_foundry_hosting import InvocationsHostServer
 from azure.identity import AzureCliCredential
 from dotenv import load_dotenv
 
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/01-basic/main.py b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/01-basic/main.py
index 047719efd9..0be716558b 100644
--- a/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/01-basic/main.py
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/01-basic/main.py
@@ -3,7 +3,8 @@
 import os
 
 from agent_framework import Agent
-from agent_framework.foundry import FoundryChatClient, ResponsesHostServer
+from agent_framework.foundry import FoundryChatClient
+from agent_framework_foundry_hosting import ResponsesHostServer
 from azure.ai.agentserver.responses import InMemoryResponseProvider
 from azure.identity import AzureCliCredential
 from dotenv import load_dotenv
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/02_local_tools/main.py b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/02_local_tools/main.py
index 6d213bf8c3..7c8b3f6dc8 100644
--- a/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/02_local_tools/main.py
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/02_local_tools/main.py
@@ -4,7 +4,8 @@
 from random import randint
 
 from agent_framework import Agent, tool
-from agent_framework.foundry import FoundryChatClient, ResponsesHostServer
+from agent_framework.foundry import FoundryChatClient
+from agent_framework_foundry_hosting import ResponsesHostServer
 from azure.ai.agentserver.responses import InMemoryResponseProvider
 from azure.identity import AzureCliCredential
 from dotenv import load_dotenv
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/03_remote_mcp/main.py b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/03_remote_mcp/main.py
index 455b13f1f2..9a244f686f 100644
--- a/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/03_remote_mcp/main.py
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/03_remote_mcp/main.py
@@ -3,7 +3,8 @@
 import os
 
 from agent_framework import Agent
-from agent_framework.foundry import FoundryChatClient, ResponsesHostServer
+from agent_framework.foundry import FoundryChatClient
+from agent_framework_foundry_hosting import ResponsesHostServer
 from azure.ai.agentserver.responses import InMemoryResponseProvider
 from azure.identity import AzureCliCredential
 from dotenv import load_dotenv
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/04-workflows/main.py b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/04-workflows/main.py
index 9d6ed68158..8965dc6a21 100644
--- a/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/04-workflows/main.py
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/04-workflows/main.py
@@ -3,8 +3,9 @@
 import os
 
 from agent_framework import Agent
-from agent_framework.foundry import FoundryChatClient, ResponsesHostServer
+from agent_framework.foundry import FoundryChatClient
 from agent_framework.orchestrations import GroupChatBuilder, GroupChatState
+from agent_framework_foundry_hosting import ResponsesHostServer
 from azure.ai.agentserver.responses import InMemoryResponseProvider
 from azure.identity import AzureCliCredential
 from dotenv import load_dotenv

From f26208757e1a5fee35fa589864585c705c24fe83 Mon Sep 17 00:00:00 2001
From: Tao Chen <taochen@microsoft.com>
Date: Thu, 9 Apr 2026 11:46:26 -0700
Subject: [PATCH 07/13] Comments 2

---
 .../core/agent_framework/_telemetry.py        |  31 +-
 python/packages/foundry_hosting/README.md     |  11 +
 .../_invocations.py                           |   4 +
 .../_message_converter.py                     | 159 ----------
 .../_responses.py                             | 284 ++++++++++++++----
 .../packages/foundry_hosting/pyproject.toml   |   7 +-
 python/pyproject.toml                         |   3 -
 7 files changed, 269 insertions(+), 230 deletions(-)
 delete mode 100644 python/packages/foundry_hosting/agent_framework_foundry_hosting/_message_converter.py

diff --git a/python/packages/core/agent_framework/_telemetry.py b/python/packages/core/agent_framework/_telemetry.py
index a044fc9d02..f4d259806d 100644
--- a/python/packages/core/agent_framework/_telemetry.py
+++ b/python/packages/core/agent_framework/_telemetry.py
@@ -26,6 +26,28 @@
 HTTP_USER_AGENT: Final[str] = "agent-framework-python"
 AGENT_FRAMEWORK_USER_AGENT = f"{HTTP_USER_AGENT}/{version_info}"  # type: ignore[has-type]
 
+_user_agent_prefixes: list[str] = []
+
+
+def append_to_user_agent(prefix: str) -> None:
+    """Prepend a prefix to the agent framework user agent string.
+
+    This is useful for hosting layers that want to identify themselves in telemetry.
+    Duplicate prefixes are ignored.
+
+    Args:
+        prefix: The prefix to prepend (e.g. "foundry-hosting-responses").
+    """
+    if prefix and prefix not in _user_agent_prefixes:
+        _user_agent_prefixes.append(prefix)
+
+
+def _get_user_agent() -> str:
+    """Return the full user agent string including any prepended prefixes."""
+    if not _user_agent_prefixes:
+        return AGENT_FRAMEWORK_USER_AGENT
+    return f"{'/'.join(_user_agent_prefixes)}/{AGENT_FRAMEWORK_USER_AGENT}"
+
 
 def prepend_agent_framework_to_user_agent(headers: dict[str, Any] | None = None) -> dict[str, Any]:
     """Prepend "agent-framework" to the User-Agent in the headers.
@@ -57,12 +79,9 @@ def prepend_agent_framework_to_user_agent(headers: dict[str, Any] | None = None)
     """
     if not IS_TELEMETRY_ENABLED:
         return headers or {}
+    user_agent = _get_user_agent()
     if not headers:
-        return {USER_AGENT_KEY: AGENT_FRAMEWORK_USER_AGENT}
-    headers[USER_AGENT_KEY] = (
-        f"{AGENT_FRAMEWORK_USER_AGENT} {headers[USER_AGENT_KEY]}"
-        if USER_AGENT_KEY in headers
-        else AGENT_FRAMEWORK_USER_AGENT
-    )
+        return {USER_AGENT_KEY: user_agent}
+    headers[USER_AGENT_KEY] = f"{user_agent} {headers[USER_AGENT_KEY]}" if USER_AGENT_KEY in headers else user_agent
 
     return headers
diff --git a/python/packages/foundry_hosting/README.md b/python/packages/foundry_hosting/README.md
index e69de29bb2..9222c01aa8 100644
--- a/python/packages/foundry_hosting/README.md
+++ b/python/packages/foundry_hosting/README.md
@@ -0,0 +1,11 @@
+# Foundry Hosting
+
+This package provides the integration of Agent Framework agents and workflows with the Foundry Agent Server, which can be hosted on Foundry infrastructure.
+
+## Responses
+
+TODO
+
+## Invocations
+
+TODO
diff --git a/python/packages/foundry_hosting/agent_framework_foundry_hosting/_invocations.py b/python/packages/foundry_hosting/agent_framework_foundry_hosting/_invocations.py
index ae3330fff4..2feb1001ba 100644
--- a/python/packages/foundry_hosting/agent_framework_foundry_hosting/_invocations.py
+++ b/python/packages/foundry_hosting/agent_framework_foundry_hosting/_invocations.py
@@ -1,6 +1,7 @@
 # Copyright (c) Microsoft. All rights reserved.
 
 from agent_framework import AgentSession, BaseAgent, SupportsAgentRun
+from agent_framework._telemetry import append_to_user_agent
 from azure.ai.agentserver.invocations import InvocationAgentServerHost
 from starlette.requests import Request
 from starlette.responses import JSONResponse, Response, StreamingResponse
@@ -10,6 +11,8 @@
 class InvocationsHostServer(InvocationAgentServerHost):
     """An invocations server host for an agent."""
 
+    USER_AGENT_PREFIX = "foundry-hosting-invocations"
+
     def __init__(
         self,
         agent: BaseAgent,
@@ -35,6 +38,7 @@ def __init__(
         if not isinstance(agent, SupportsAgentRun):
             raise TypeError("Agent must support the SupportsAgentRun interface")
 
+        append_to_user_agent(self.USER_AGENT_PREFIX)
         self._agent = agent
         self._stream = stream
         self._sessions: dict[str, AgentSession] = {}
diff --git a/python/packages/foundry_hosting/agent_framework_foundry_hosting/_message_converter.py b/python/packages/foundry_hosting/agent_framework_foundry_hosting/_message_converter.py
deleted file mode 100644
index f76306af11..0000000000
--- a/python/packages/foundry_hosting/agent_framework_foundry_hosting/_message_converter.py
+++ /dev/null
@@ -1,159 +0,0 @@
-# Copyright (c) Microsoft. All rights reserved.
-
-from agent_framework import Content, Message
-from azure.ai.agentserver.responses.models import (
-    ComputerScreenshotContent,
-    FunctionCallOutputItemParam,
-    MessageContent,
-    MessageContentInputFileContent,
-    MessageContentInputImageContent,
-    MessageContentInputTextContent,
-    MessageContentOutputTextContent,
-    MessageContentReasoningTextContent,
-    MessageContentRefusalContent,
-    OutputItem,
-    OutputItemFunctionToolCall,
-    OutputItemMessage,
-    OutputItemOutputMessage,
-    OutputItemReasoningItem,
-    OutputMessageContent,
-    OutputMessageContentOutputTextContent,
-    OutputMessageContentRefusalContent,
-    SummaryTextContent,
-    TextContent,
-)
-from typing_extensions import Sequence, cast
-
-
-def to_messages(history: Sequence[OutputItem]) -> list[Message]:
-    """Converts a sequence of OutputItem objects to a list of Message objects.
-
-    Args:
-        history (Sequence[OutputItem]): The sequence of OutputItem objects to convert.
-
-    Returns:
-        list[Message]: The list of Message objects.
-    """
-    messages: list[Message] = []
-    for item in history:
-        messages.append(_to_message(item))
-    return messages
-
-
-def _to_message(item: OutputItem) -> Message:
-    """Converts an OutputItem to a Message.
-
-    Args:
-        item (OutputItem): The OutputItem to convert.
-
-    Returns:
-        Message: The converted Message.
-
-    Raises:
-        ValueError: If the OutputItem type is not supported.
-    """
-    if item.type == "output_message":
-        msg = cast(OutputItemOutputMessage, item)
-        contents = [_convert_output_message_content(part) for part in msg.content]
-        return Message(role=msg.role, contents=contents)
-
-    if item.type == "message":
-        msg = cast(OutputItemMessage, item)
-        contents = [_convert_message_content(part) for part in msg.content]
-        return Message(role=msg.role, contents=contents)
-
-    if item.type == "function_call":
-        fc = cast(OutputItemFunctionToolCall, item)
-        return Message(
-            role="assistant",
-            contents=[Content.from_function_call(fc.call_id, fc.name, arguments=fc.arguments)],
-        )
-
-    if item.type == "function_call_output":
-        fco = cast(FunctionCallOutputItemParam, item)
-        output = fco.output if isinstance(fco.output, str) else str(fco.output)
-        return Message(
-            role="tool",
-            contents=[Content.from_function_result(fco.call_id, result=output)],
-        )
-
-    if item.type == "reasoning":
-        reasoning = cast(OutputItemReasoningItem, item)
-        contents: list[Content] = []
-        if reasoning.summary:
-            for summary in reasoning.summary:
-                contents.append(Content.from_text(summary.text))
-        return Message(role="assistant", contents=contents)
-
-    raise ValueError(f"Unsupported OutputItem type: {item.type}")
-
-
-def _convert_output_message_content(content: OutputMessageContent) -> Content:
-    """Converts an OutputMessageContent to a Content object.
-
-    Args:
-        content (OutputMessageContent): The OutputMessageContent to convert.
-
-    Returns:
-        Content: The converted Content object.
-
-    Raises:
-        ValueError: If the OutputMessageContent type is not supported.
-    """
-    if content.type == "output_text":
-        text_content = cast(OutputMessageContentOutputTextContent, content)
-        return Content.from_text(text_content.text)
-    if content.type == "refusal":
-        refusal_content = cast(OutputMessageContentRefusalContent, content)
-        return Content.from_text(refusal_content.refusal)
-
-    raise ValueError(f"Unsupported OutputMessageContent type: {content.type}")
-
-
-def _convert_message_content(content: MessageContent) -> Content:
-    """Converts a MessageContent to a Content object.
-
-    Args:
-        content (MessageContent): The MessageContent to convert.
-
-    Returns:
-        Content: The converted Content object.
-
-    Raises:
-        ValueError: If the MessageContent type is not supported.
-    """
-    if content.type == "input_text":
-        input_text = cast(MessageContentInputTextContent, content)
-        return Content.from_text(input_text.text)
-    if content.type == "output_text":
-        output_text = cast(MessageContentOutputTextContent, content)
-        return Content.from_text(output_text.text)
-    if content.type == "text":
-        text = cast(TextContent, content)
-        return Content.from_text(text.text)
-    if content.type == "summary_text":
-        summary = cast(SummaryTextContent, content)
-        return Content.from_text(summary.text)
-    if content.type == "refusal":
-        refusal = cast(MessageContentRefusalContent, content)
-        return Content.from_text(refusal.refusal)
-    if content.type == "reasoning_text":
-        reasoning = cast(MessageContentReasoningTextContent, content)
-        return Content.from_text_reasoning(text=reasoning.text)
-    if content.type == "input_image":
-        image = cast(MessageContentInputImageContent, content)
-        if image.image_url:
-            return Content.from_uri(image.image_url)
-        if image.file_id:
-            return Content.from_hosted_file(image.file_id)
-    if content.type == "input_file":
-        file = cast(MessageContentInputFileContent, content)
-        if file.file_url:
-            return Content.from_uri(file.file_url)
-        if file.file_id:
-            return Content.from_hosted_file(file.file_id, name=file.filename)
-    if content.type == "computer_screenshot":
-        screenshot = cast(ComputerScreenshotContent, content)
-        return Content.from_uri(screenshot.image_url)
-
-    raise ValueError(f"Unsupported MessageContent type: {content.type}")
diff --git a/python/packages/foundry_hosting/agent_framework_foundry_hosting/_responses.py b/python/packages/foundry_hosting/agent_framework_foundry_hosting/_responses.py
index 221ae279ca..99c542e4e4 100644
--- a/python/packages/foundry_hosting/agent_framework_foundry_hosting/_responses.py
+++ b/python/packages/foundry_hosting/agent_framework_foundry_hosting/_responses.py
@@ -3,7 +3,8 @@
 import asyncio
 from collections.abc import AsyncIterable
 
-from agent_framework import BaseAgent, HistoryProvider, Message, SupportsAgentRun
+from agent_framework import Agent, ChatOptions, Content, HistoryProvider, Message
+from agent_framework._telemetry import append_to_user_agent
 from azure.ai.agentserver.responses import (
     ResponseContext,
     ResponseEventStream,
@@ -11,51 +12,40 @@
     ResponsesServerOptions,
 )
 from azure.ai.agentserver.responses.hosting import ResponsesAgentServerHost
-from azure.ai.agentserver.responses.models import CreateResponse, get_input_text
-from typing_extensions import Any, Sequence
-
-from ._message_converter import to_messages
-
-
-class ResponsesHostHistoryProvider(HistoryProvider):
-    """A history provider that retrieves messages from a ResponseContext."""
-
-    def __init__(self, context: ResponseContext):
-        """Initialize a ResponsesHostHistoryProvider.
-
-        Args:
-            context: The ResponseContext to retrieve messages from.
-        """
-        super().__init__("responses-host", load_messages=True)
-        self.context = context
-
-    async def get_messages(
-        self,
-        session_id: str | None,
-        *,
-        state: dict[str, Any] | None = None,
-        **kwargs: Any,
-    ) -> list[Message]:
-        history = await self.context.get_history()
-        return to_messages(history)
-
-    async def save_messages(
-        self,
-        session_id: str | None,
-        messages: Sequence[Message],
-        *,
-        state: dict[str, Any] | None = None,
-        **kwargs: Any,
-    ) -> None:
-        pass
+from azure.ai.agentserver.responses.models import (
+    ComputerScreenshotContent,
+    CreateResponse,
+    FunctionCallOutputItemParam,
+    MessageContent,
+    MessageContentInputFileContent,
+    MessageContentInputImageContent,
+    MessageContentInputTextContent,
+    MessageContentOutputTextContent,
+    MessageContentReasoningTextContent,
+    MessageContentRefusalContent,
+    OutputItem,
+    OutputItemFunctionToolCall,
+    OutputItemMessage,
+    OutputItemOutputMessage,
+    OutputItemReasoningItem,
+    OutputMessageContent,
+    OutputMessageContentOutputTextContent,
+    OutputMessageContentRefusalContent,
+    SummaryTextContent,
+    TextContent,
+    get_input_text,
+)
+from typing_extensions import Any, Sequence, cast
 
 
 class ResponsesHostServer(ResponsesAgentServerHost):
     """A responses server host for an agent."""
 
+    USER_AGENT_PREFIX = "foundry-hosting-responses"
+
     def __init__(
         self,
-        agent: BaseAgent,
+        agent: Agent,
         *,
         prefix: str = "",
         options: ResponsesServerOptions | None = None,
@@ -72,39 +62,42 @@ def __init__(
             **kwargs: Additional keyword arguments.
 
         Note:
-            If the agent has a history provider with `load_messages=True`, it will be
-            replaced with a `ResponsesHostHistoryProvider` that will retrieve history
-            from the hosting infrastructure.
+            The agent must not have a history provider with `load_messages=True`,
+            because history is managed by the hosting infrastructure.
         """
         super().__init__(prefix=prefix, options=options, provider=provider, **kwargs)
 
-        if not isinstance(agent, SupportsAgentRun):
-            raise TypeError("Agent must support the SupportsAgentRun interface")
-
+        self._validate_agent(agent)
         self._agent = agent
         self.create_handler(self._handle_create)  # pyright: ignore[reportUnknownMemberType]
 
+        # Append the user agent prefix for telemetry purposes
+        append_to_user_agent(self.USER_AGENT_PREFIX)
+
+    def _validate_agent(self, agent: Agent) -> None:
+        """Validate the agent to ensure it does not have a history provider with `load_messages=True`.
+
+        History is managed by the hosting infrastructure.
+        """
+        for provider in agent.context_providers:
+            if isinstance(provider, HistoryProvider) and provider.load_messages:
+                raise RuntimeError(
+                    "There shouldn't be a history provider with `load_messages=True` already present. "
+                    "History is managed by the hosting infrastructure."
+                )
+
     async def _handle_create(
         self,
         request: CreateResponse,
         context: ResponseContext,
         cancellation_signal: asyncio.Event,
     ) -> AsyncIterable[dict[str, Any]]:
-        # Replace or add a history provider that has `load_messages=True`
-        history_provider_idx: list[int] = []
-        for i, provider in enumerate(self._agent.context_providers):
-            if isinstance(provider, HistoryProvider) and provider.load_messages:
-                history_provider_idx.append(i)
-
-        if not history_provider_idx:
-            self._agent.context_providers.append(ResponsesHostHistoryProvider(context))
-        elif len(history_provider_idx) > 1:
-            # There shouldn't be more than one history provider with `load_messages=True`
-            raise RuntimeError("There shouldn't be more than one history provider with `load_messages=True`")
-        else:
-            self._agent.context_providers[history_provider_idx[0]] = ResponsesHostHistoryProvider(context)
-
+        """Handle the creation of a response."""
         input_items = get_input_text(request)
+        history = await context.get_history()
+        messages = [*_to_messages(history), input_items]
+
+        chat_options = _to_chat_options(request)
 
         stream = ResponseEventStream(response_id=context.response_id, model=request.model)
 
@@ -115,7 +108,7 @@ async def _handle_create(
 
         if request.stream is None or request.stream is False:
             # Run the agent in non-streaming mode
-            response = await self._agent.run(input_items, stream=False)
+            response = await self._agent.run(messages, stream=False, options=chat_options)
             for item in stream.output_item_message(response.text):
                 yield item
             yield stream.emit_completed()
@@ -128,7 +121,7 @@ async def _handle_create(
 
         # Invoke the MAF agent
         full_text = ""
-        async for update in self._agent.run(input_items, stream=True):
+        async for update in self._agent.run(messages, stream=True, options=chat_options):
             full_text += update.text
             yield text_content.emit_delta(update.text)
 
@@ -138,3 +131,172 @@ async def _handle_create(
         yield message_item.emit_done()
 
         yield stream.emit_completed()
+
+
+# region Option Conversion
+
+
+def _to_chat_options(request: CreateResponse) -> ChatOptions:
+    """Converts a CreateResponse request to ChatOptions.
+
+    Args:
+        request (CreateResponse): The request to convert.
+
+    Returns:
+        ChatOptions: The converted ChatOptions.
+    """
+    chat_options = ChatOptions()
+
+    if request.temperature is not None:
+        chat_options["temperature"] = request.temperature
+    if request.top_p is not None:
+        chat_options["top_p"] = request.top_p
+    if request.max_output_tokens is not None:
+        chat_options["max_tokens"] = request.max_output_tokens
+    if request.parallel_tool_calls is not None:
+        chat_options["allow_multiple_tool_calls"] = request.parallel_tool_calls
+
+    return chat_options
+
+
+# endregion
+
+
+# region Message Conversion
+
+
+def _to_messages(history: Sequence[OutputItem]) -> list[Message]:
+    """Converts a sequence of OutputItem objects to a list of Message objects.
+
+    Args:
+        history (Sequence[OutputItem]): The sequence of OutputItem objects to convert.
+
+    Returns:
+        list[Message]: The list of Message objects.
+    """
+    messages: list[Message] = []
+    for item in history:
+        messages.append(_to_message(item))
+    return messages
+
+
+def _to_message(item: OutputItem) -> Message:
+    """Converts an OutputItem to a Message.
+
+    Args:
+        item (OutputItem): The OutputItem to convert.
+
+    Returns:
+        Message: The converted Message.
+
+    Raises:
+        ValueError: If the OutputItem type is not supported.
+    """
+    if item.type == "output_message":
+        msg = cast(OutputItemOutputMessage, item)
+        contents = [_convert_output_message_content(part) for part in msg.content]
+        return Message(role=msg.role, contents=contents)
+
+    if item.type == "message":
+        msg = cast(OutputItemMessage, item)
+        contents = [_convert_message_content(part) for part in msg.content]
+        return Message(role=msg.role, contents=contents)
+
+    if item.type == "function_call":
+        fc = cast(OutputItemFunctionToolCall, item)
+        return Message(
+            role="assistant",
+            contents=[Content.from_function_call(fc.call_id, fc.name, arguments=fc.arguments)],
+        )
+
+    if item.type == "function_call_output":
+        fco = cast(FunctionCallOutputItemParam, item)
+        output = fco.output if isinstance(fco.output, str) else str(fco.output)
+        return Message(
+            role="tool",
+            contents=[Content.from_function_result(fco.call_id, result=output)],
+        )
+
+    if item.type == "reasoning":
+        reasoning = cast(OutputItemReasoningItem, item)
+        contents: list[Content] = []
+        if reasoning.summary:
+            for summary in reasoning.summary:
+                contents.append(Content.from_text(summary.text))
+        return Message(role="assistant", contents=contents)
+
+    raise ValueError(f"Unsupported OutputItem type: {item.type}")
+
+
+def _convert_output_message_content(content: OutputMessageContent) -> Content:
+    """Converts an OutputMessageContent to a Content object.
+
+    Args:
+        content (OutputMessageContent): The OutputMessageContent to convert.
+
+    Returns:
+        Content: The converted Content object.
+
+    Raises:
+        ValueError: If the OutputMessageContent type is not supported.
+    """
+    if content.type == "output_text":
+        text_content = cast(OutputMessageContentOutputTextContent, content)
+        return Content.from_text(text_content.text)
+    if content.type == "refusal":
+        refusal_content = cast(OutputMessageContentRefusalContent, content)
+        return Content.from_text(refusal_content.refusal)
+
+    raise ValueError(f"Unsupported OutputMessageContent type: {content.type}")
+
+
+def _convert_message_content(content: MessageContent) -> Content:
+    """Converts a MessageContent to a Content object.
+
+    Args:
+        content (MessageContent): The MessageContent to convert.
+
+    Returns:
+        Content: The converted Content object.
+
+    Raises:
+        ValueError: If the MessageContent type is not supported.
+    """
+    if content.type == "input_text":
+        input_text = cast(MessageContentInputTextContent, content)
+        return Content.from_text(input_text.text)
+    if content.type == "output_text":
+        output_text = cast(MessageContentOutputTextContent, content)
+        return Content.from_text(output_text.text)
+    if content.type == "text":
+        text = cast(TextContent, content)
+        return Content.from_text(text.text)
+    if content.type == "summary_text":
+        summary = cast(SummaryTextContent, content)
+        return Content.from_text(summary.text)
+    if content.type == "refusal":
+        refusal = cast(MessageContentRefusalContent, content)
+        return Content.from_text(refusal.refusal)
+    if content.type == "reasoning_text":
+        reasoning = cast(MessageContentReasoningTextContent, content)
+        return Content.from_text_reasoning(text=reasoning.text)
+    if content.type == "input_image":
+        image = cast(MessageContentInputImageContent, content)
+        if image.image_url:
+            return Content.from_uri(image.image_url)
+        if image.file_id:
+            return Content.from_hosted_file(image.file_id)
+    if content.type == "input_file":
+        file = cast(MessageContentInputFileContent, content)
+        if file.file_url:
+            return Content.from_uri(file.file_url)
+        if file.file_id:
+            return Content.from_hosted_file(file.file_id, name=file.filename)
+    if content.type == "computer_screenshot":
+        screenshot = cast(ComputerScreenshotContent, content)
+        return Content.from_uri(screenshot.image_url)
+
+    raise ValueError(f"Unsupported MessageContent type: {content.type}")
+
+
+# endregion
diff --git a/python/packages/foundry_hosting/pyproject.toml b/python/packages/foundry_hosting/pyproject.toml
index 3bb6c71495..55916ac076 100644
--- a/python/packages/foundry_hosting/pyproject.toml
+++ b/python/packages/foundry_hosting/pyproject.toml
@@ -4,7 +4,7 @@ description = "Foundry Hosting integration for Microsoft Agent Framework."
 authors = [{ name = "Microsoft", email = "af-support@microsoft.com"}]
 readme = "README.md"
 requires-python = ">=3.10"
-version = "1.0.0b260402"
+version = "1.0.0a260402"
 license-files = ["LICENSE"]
 urls.homepage = "https://aka.ms/agent-framework"
 urls.source = "https://github.com/microsoft/agent-framework/tree/main/python"
@@ -29,6 +29,11 @@ dependencies = [
     "azure-ai-agentserver-invocations"
 ]
 
+[tool.uv.sources]
+azure-ai-agentserver-responses = { git = "https://github.com/Azure/azure-sdk-for-python.git", branch = "agentserver/responses", subdirectory = "sdk/agentserver/azure-ai-agentserver-responses" }
+azure-ai-agentserver-invocations = { git = "https://github.com/Azure/azure-sdk-for-python.git", branch = "agentserver/responses", subdirectory = "sdk/agentserver/azure-ai-agentserver-invocations" }
+azure-ai-agentserver-core = { git = "https://github.com/Azure/azure-sdk-for-python.git", branch = "agentserver/responses", subdirectory = "sdk/agentserver/azure-ai-agentserver-core" }
+
 [tool.uv]
 prerelease = "if-necessary-or-explicit"
 environments = [
diff --git a/python/pyproject.toml b/python/pyproject.toml
index 4accd6d888..9e677fb602 100644
--- a/python/pyproject.toml
+++ b/python/pyproject.toml
@@ -89,9 +89,6 @@ agent-framework-redis = { workspace = true }
 agent-framework-github-copilot = { workspace = true }
 agent-framework-claude = { workspace = true }
 agent-framework-orchestrations = { workspace = true }
-azure-ai-agentserver-responses = { git = "https://github.com/Azure/azure-sdk-for-python.git", branch = "agentserver/responses", subdirectory = "sdk/agentserver/azure-ai-agentserver-responses" }
-azure-ai-agentserver-invocations = { git = "https://github.com/Azure/azure-sdk-for-python.git", branch = "agentserver/responses", subdirectory = "sdk/agentserver/azure-ai-agentserver-invocations" }
-azure-ai-agentserver-core = { git = "https://github.com/Azure/azure-sdk-for-python.git", branch = "agentserver/responses", subdirectory = "sdk/agentserver/azure-ai-agentserver-core" }
 litellm = { url = "https://files.pythonhosted.org/packages/57/77/0c6eca2cb049793ddf8ce9cdcd5123a35666c4962514788c4fc90edf1d3b/litellm-1.82.1-py3-none-any.whl" }
 
 [tool.ruff]

From 228c60ca0b57fe61c38cce4d2762a608320447b6 Mon Sep 17 00:00:00 2001
From: Tao Chen <taochen@microsoft.com>
Date: Thu, 9 Apr 2026 12:25:39 -0700
Subject: [PATCH 08/13] Comments 3

---
 .../agent_framework_foundry_hosting/_responses.py    | 12 +++++++-----
 1 file changed, 7 insertions(+), 5 deletions(-)

diff --git a/python/packages/foundry_hosting/agent_framework_foundry_hosting/_responses.py b/python/packages/foundry_hosting/agent_framework_foundry_hosting/_responses.py
index 99c542e4e4..8c0447d7bd 100644
--- a/python/packages/foundry_hosting/agent_framework_foundry_hosting/_responses.py
+++ b/python/packages/foundry_hosting/agent_framework_foundry_hosting/_responses.py
@@ -112,6 +112,7 @@ async def _handle_create(
             for item in stream.output_item_message(response.text):
                 yield item
             yield stream.emit_completed()
+            return
 
         # Start the streaming response
         message_item = stream.add_output_item_message()
@@ -120,13 +121,14 @@ async def _handle_create(
         yield text_content.emit_added()
 
         # Invoke the MAF agent
-        full_text = ""
-        async for update in self._agent.run(messages, stream=True, options=chat_options):
-            full_text += update.text
-            yield text_content.emit_delta(update.text)
+        response_stream = self._agent.run(messages, stream=True, options=chat_options)
+        async for update in response_stream:
+            if update.text:
+                yield text_content.emit_delta(update.text)
 
         # Complete the message
-        yield text_content.emit_done(full_text)
+        final = await response_stream.get_final_response()
+        yield text_content.emit_done(final.text)
         yield message_item.emit_content_done(text_content)
         yield message_item.emit_done()
 

From 735c1236ee219629cadd3a216a6417c7370676ab Mon Sep 17 00:00:00 2001
From: Tao Chen <taochen@microsoft.com>
Date: Thu, 9 Apr 2026 12:46:42 -0700
Subject: [PATCH 09/13] Improve README

---
 .../foundry_hosted_responses/01-basic/README.md  | 16 ++++++++++++++++
 1 file changed, 16 insertions(+)

diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/01-basic/README.md b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/01-basic/README.md
index 57686cddda..e6eaa7c6c7 100644
--- a/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/01-basic/README.md
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/01-basic/README.md
@@ -1,19 +1,35 @@
 # Basic example of hosting an agent with the `responses` API
 
+## Running the server locally
+
 Run the following command to start the server:
 
 ```bash
 python main.py
 ```
 
+## Interacting with the agent
+
 Send a POST request to the server with a JSON body containing a "message" field to interact with the agent. For example:
 
 ```bash
 curl -X POST http://localhost:8088/responses -H "Content-Type: application/json" -d '{"input": "Hi"}'
 ```
 
+The server will respond with a JSON object containing the response text and a response ID. You can use this response ID to continue the conversation in subsequent requests.
+
+## Multi-turn conversation
+
 To have a multi-turn conversation with the agent, include the previous response id in the request body. For example:
 
 ```bash
 curl -X POST http://localhost:8088/responses -H "Content-Type: application/json" -d '{"input": "How are you?", "previous_response_id": "REPLACE_WITH_PREVIOUS_RESPONSE_ID"}'
 ```
+
+## Deploying to Foundry
+
+TODO
+
+## Using the deployed agent in Agent Framework
+
+TODO

From 20ac97c69d183d51c5c28bcc9fa51ae2271055e2 Mon Sep 17 00:00:00 2001
From: Tao Chen <taochen@microsoft.com>
Date: Thu, 9 Apr 2026 14:15:34 -0700
Subject: [PATCH 10/13] Add local shell sample

---
 .../{01-basic => 01_basic}/.dockerignore      |  0
 .../{01-basic => 01_basic}/Dockerfile         |  0
 .../{01-basic => 01_basic}/README.md          |  0
 .../agent.manifest.yaml                       |  0
 .../{01-basic => 01_basic}/agent.yaml         |  0
 .../{01-basic => 01_basic}/main.py            |  0
 .../{01-basic => 01_basic}/requirements.txt   |  0
 .../.dockerignore                             |  0
 .../{04-workflows => 04_workflows}/Dockerfile |  0
 .../{04-workflows => 04_workflows}/README.md  |  0
 .../agent.manifest.yaml                       |  0
 .../{04-workflows => 04_workflows}/agent.yaml |  0
 .../{04-workflows => 04_workflows}/main.py    |  0
 .../requirements.txt                          |  0
 .../05_local_shell/.dockerignore              |  6 ++
 .../05_local_shell/Dockerfile                 | 16 +++++
 .../05_local_shell/README.md                  | 43 +++++++++++++
 .../05_local_shell/agent.manifest.yaml        | 15 +++++
 .../05_local_shell/agent.yaml                 |  8 +++
 .../05_local_shell/main.py                    | 63 +++++++++++++++++++
 .../05_local_shell/requirements.txt           |  2 +
 21 files changed, 153 insertions(+)
 rename python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/{01-basic => 01_basic}/.dockerignore (100%)
 rename python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/{01-basic => 01_basic}/Dockerfile (100%)
 rename python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/{01-basic => 01_basic}/README.md (100%)
 rename python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/{01-basic => 01_basic}/agent.manifest.yaml (100%)
 rename python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/{01-basic => 01_basic}/agent.yaml (100%)
 rename python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/{01-basic => 01_basic}/main.py (100%)
 rename python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/{01-basic => 01_basic}/requirements.txt (100%)
 rename python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/{04-workflows => 04_workflows}/.dockerignore (100%)
 rename python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/{04-workflows => 04_workflows}/Dockerfile (100%)
 rename python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/{04-workflows => 04_workflows}/README.md (100%)
 rename python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/{04-workflows => 04_workflows}/agent.manifest.yaml (100%)
 rename python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/{04-workflows => 04_workflows}/agent.yaml (100%)
 rename python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/{04-workflows => 04_workflows}/main.py (100%)
 rename python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/{04-workflows => 04_workflows}/requirements.txt (100%)
 create mode 100644 python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/05_local_shell/.dockerignore
 create mode 100644 python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/05_local_shell/Dockerfile
 create mode 100644 python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/05_local_shell/README.md
 create mode 100644 python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/05_local_shell/agent.manifest.yaml
 create mode 100644 python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/05_local_shell/agent.yaml
 create mode 100644 python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/05_local_shell/main.py
 create mode 100644 python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/05_local_shell/requirements.txt

diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/01-basic/.dockerignore b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/01_basic/.dockerignore
similarity index 100%
rename from python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/01-basic/.dockerignore
rename to python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/01_basic/.dockerignore
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/01-basic/Dockerfile b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/01_basic/Dockerfile
similarity index 100%
rename from python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/01-basic/Dockerfile
rename to python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/01_basic/Dockerfile
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/01-basic/README.md b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/01_basic/README.md
similarity index 100%
rename from python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/01-basic/README.md
rename to python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/01_basic/README.md
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/01-basic/agent.manifest.yaml b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/01_basic/agent.manifest.yaml
similarity index 100%
rename from python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/01-basic/agent.manifest.yaml
rename to python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/01_basic/agent.manifest.yaml
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/01-basic/agent.yaml b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/01_basic/agent.yaml
similarity index 100%
rename from python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/01-basic/agent.yaml
rename to python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/01_basic/agent.yaml
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/01-basic/main.py b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/01_basic/main.py
similarity index 100%
rename from python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/01-basic/main.py
rename to python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/01_basic/main.py
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/01-basic/requirements.txt b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/01_basic/requirements.txt
similarity index 100%
rename from python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/01-basic/requirements.txt
rename to python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/01_basic/requirements.txt
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/04-workflows/.dockerignore b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/04_workflows/.dockerignore
similarity index 100%
rename from python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/04-workflows/.dockerignore
rename to python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/04_workflows/.dockerignore
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/04-workflows/Dockerfile b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/04_workflows/Dockerfile
similarity index 100%
rename from python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/04-workflows/Dockerfile
rename to python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/04_workflows/Dockerfile
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/04-workflows/README.md b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/04_workflows/README.md
similarity index 100%
rename from python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/04-workflows/README.md
rename to python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/04_workflows/README.md
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/04-workflows/agent.manifest.yaml b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/04_workflows/agent.manifest.yaml
similarity index 100%
rename from python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/04-workflows/agent.manifest.yaml
rename to python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/04_workflows/agent.manifest.yaml
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/04-workflows/agent.yaml b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/04_workflows/agent.yaml
similarity index 100%
rename from python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/04-workflows/agent.yaml
rename to python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/04_workflows/agent.yaml
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/04-workflows/main.py b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/04_workflows/main.py
similarity index 100%
rename from python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/04-workflows/main.py
rename to python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/04_workflows/main.py
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/04-workflows/requirements.txt b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/04_workflows/requirements.txt
similarity index 100%
rename from python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/04-workflows/requirements.txt
rename to python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/04_workflows/requirements.txt
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/05_local_shell/.dockerignore b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/05_local_shell/.dockerignore
new file mode 100644
index 0000000000..008e6e6616
--- /dev/null
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/05_local_shell/.dockerignore
@@ -0,0 +1,6 @@
+.venv
+__pycache__
+*.pyc
+*.pyo
+*.pyd
+.Python
\ No newline at end of file
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/05_local_shell/Dockerfile b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/05_local_shell/Dockerfile
new file mode 100644
index 0000000000..845d325e7c
--- /dev/null
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/05_local_shell/Dockerfile
@@ -0,0 +1,16 @@
+FROM python:3.12-slim
+
+WORKDIR /app/user_agent
+
+COPY wheels/ /tmp/wheels/
+COPY requirements.txt .
+RUN pip install --no-cache-dir --find-links /tmp/wheels/ -r requirements.txt && rm -rf /tmp/wheels/
+
+COPY . .
+
+RUN useradd -r appuser
+USER appuser
+
+EXPOSE 8088
+
+CMD ["python", "main.py"]
\ No newline at end of file
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/05_local_shell/README.md b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/05_local_shell/README.md
new file mode 100644
index 0000000000..73f1ad05a9
--- /dev/null
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/05_local_shell/README.md
@@ -0,0 +1,43 @@
+# Agent Framework Agent with Local Shell
+
+> Note: This agent can execute local shell commands. We recommend running it in an isolated environment for security reasons.
+
+## Running the server in a Docker container
+
+Build the Docker image:
+
+```bash
+docker build -t agent-framework-agent-with-local-shell .
+```
+
+Run the Docker container:
+
+```bash
+docker run -p 8088:8088 --env-file .env agent-framework-agent-with-local-shell
+```
+
+## Interacting with the agent
+
+Send a POST request to the server with a JSON body containing a "message" field to interact with the agent. For example:
+
+```bash
+curl -X POST http://localhost:8088/responses -H "Content-Type: application/json" -d '{"input": "Hi"}'
+```
+
+The server will respond with a JSON object containing the response text and a response ID. You can use this response ID to continue the conversation in subsequent requests.
+
+## Multi-turn conversation
+
+To have a multi-turn conversation with the agent, include the previous response id in the request body. For example:
+
+```bash
+curl -X POST http://localhost:8088/responses -H "Content-Type: application/json" -d '{"input": "How are you?", "previous_response_id": "REPLACE_WITH_PREVIOUS_RESPONSE_ID"}'
+```
+
+## Deploying to Foundry
+
+TODO
+
+## Using the deployed agent in Agent Framework
+
+TODO
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/05_local_shell/agent.manifest.yaml b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/05_local_shell/agent.manifest.yaml
new file mode 100644
index 0000000000..32f133d12d
--- /dev/null
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/05_local_shell/agent.manifest.yaml
@@ -0,0 +1,15 @@
+name: agent-framework-agent-with-local-shell
+description: >
+  An Agent Framework agent that can execute local shell commands hosted by Foundry.
+metadata:
+  tags:
+    - AI Agent Hosting
+    - Azure AI AgentServer
+    - Responses Protocol
+    - Streaming
+template:
+  name: agent-framework-agent-with-local-shell
+  kind: hosted
+  protocols:
+    - protocol: responses
+      version: v0.1.0
\ No newline at end of file
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/05_local_shell/agent.yaml b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/05_local_shell/agent.yaml
new file mode 100644
index 0000000000..6c0a04b83b
--- /dev/null
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/05_local_shell/agent.yaml
@@ -0,0 +1,8 @@
+kind: hosted
+name: agent-framework-agent-with-local-shell
+protocols:
+  - protocol: responses
+    version: v0.1.0
+resources:
+  cpu: "0.25"
+  memory: 0.5Gi
\ No newline at end of file
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/05_local_shell/main.py b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/05_local_shell/main.py
new file mode 100644
index 0000000000..094eb766f4
--- /dev/null
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/05_local_shell/main.py
@@ -0,0 +1,63 @@
+# Copyright (c) Microsoft. All rights reserved.
+
+import os
+import subprocess
+
+from agent_framework import Agent, tool
+from agent_framework.foundry import FoundryChatClient
+from agent_framework_foundry_hosting import ResponsesHostServer
+from azure.ai.agentserver.responses import InMemoryResponseProvider
+from azure.identity import AzureCliCredential
+from dotenv import load_dotenv
+
+# Load environment variables from .env file
+load_dotenv()
+
+
+@tool(approval_mode="always_require")
+def run_bash(command: str) -> str:
+    """Execute a shell command locally and return stdout, stderr, and exit code."""
+    try:
+        result = subprocess.run(
+            command,
+            shell=True,
+            capture_output=True,
+            text=True,
+            timeout=30,
+        )
+        parts: list[str] = []
+        if result.stdout:
+            parts.append(result.stdout)
+        if result.stderr:
+            parts.append(f"stderr: {result.stderr}")
+        parts.append(f"exit_code: {result.returncode}")
+        return "\n".join(parts)
+    except subprocess.TimeoutExpired:
+        return "Command timed out after 30 seconds"
+    except Exception as e:
+        return f"Error executing command: {e}"
+
+
+def main():
+    client = FoundryChatClient(
+        project_endpoint=os.environ["FOUNDRY_PROJECT_ENDPOINT"],
+        model=os.environ["FOUNDRY_MODEL"],
+        credential=AzureCliCredential(),
+    )
+
+    agent = Agent(
+        client=client,
+        instructions="You are a friendly assistant. Keep your answers brief.",
+        tools=[run_bash],
+        # History will be managed by the hosting infrastructure, thus there
+        # is no need to store history by the service. Learn more at:
+        # https://developers.openai.com/api/reference/resources/responses/methods/create
+        default_options={"store": False},
+    )
+
+    server = ResponsesHostServer(agent, provider=InMemoryResponseProvider())
+    server.run()
+
+
+if __name__ == "__main__":
+    main()
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/05_local_shell/requirements.txt b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/05_local_shell/requirements.txt
new file mode 100644
index 0000000000..61b5ba14f7
--- /dev/null
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/05_local_shell/requirements.txt
@@ -0,0 +1,2 @@
+agent-framework-core
+agent-framework-foundry-hosting
\ No newline at end of file

From 8ecf274abedc07ba7cf5f7182a317dd08a5f3569 Mon Sep 17 00:00:00 2001
From: Tao Chen <taochen@microsoft.com>
Date: Thu, 9 Apr 2026 17:03:00 -0700
Subject: [PATCH 11/13] WIP: Add eval and memory samples

---
 .../06_eval/.dockerignore                     |  6 ++
 .../06_eval/Dockerfile                        | 16 ++++
 .../06_eval/README.md                         | 35 +++++++
 .../06_eval/agent.manifest.yaml               | 15 +++
 .../06_eval/agent.yaml                        |  8 ++
 .../foundry_hosted_responses/06_eval/main.py  | 50 ++++++++++
 .../06_eval/requirements.txt                  |  2 +
 .../07_foundry_memory/.dockerignore           |  6 ++
 .../07_foundry_memory/Dockerfile              | 16 ++++
 .../07_foundry_memory/README.md               | 35 +++++++
 .../07_foundry_memory/agent.manifest.yaml     | 15 +++
 .../07_foundry_memory/agent.yaml              |  8 ++
 .../07_foundry_memory/main.py                 | 91 +++++++++++++++++++
 .../07_foundry_memory/requirements.txt        |  2 +
 14 files changed, 305 insertions(+)
 create mode 100644 python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/06_eval/.dockerignore
 create mode 100644 python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/06_eval/Dockerfile
 create mode 100644 python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/06_eval/README.md
 create mode 100644 python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/06_eval/agent.manifest.yaml
 create mode 100644 python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/06_eval/agent.yaml
 create mode 100644 python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/06_eval/main.py
 create mode 100644 python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/06_eval/requirements.txt
 create mode 100644 python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/07_foundry_memory/.dockerignore
 create mode 100644 python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/07_foundry_memory/Dockerfile
 create mode 100644 python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/07_foundry_memory/README.md
 create mode 100644 python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/07_foundry_memory/agent.manifest.yaml
 create mode 100644 python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/07_foundry_memory/agent.yaml
 create mode 100644 python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/07_foundry_memory/main.py
 create mode 100644 python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/07_foundry_memory/requirements.txt

diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/06_eval/.dockerignore b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/06_eval/.dockerignore
new file mode 100644
index 0000000000..008e6e6616
--- /dev/null
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/06_eval/.dockerignore
@@ -0,0 +1,6 @@
+.venv
+__pycache__
+*.pyc
+*.pyo
+*.pyd
+.Python
\ No newline at end of file
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/06_eval/Dockerfile b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/06_eval/Dockerfile
new file mode 100644
index 0000000000..eaffb94f19
--- /dev/null
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/06_eval/Dockerfile
@@ -0,0 +1,16 @@
+FROM python:3.12-slim
+
+WORKDIR /app
+
+COPY . user_agent/
+WORKDIR /app/user_agent
+
+RUN if [ -f requirements.txt ]; then \
+        pip install -r requirements.txt; \
+    else \
+        echo "No requirements.txt found"; \
+    fi
+
+EXPOSE 8088
+
+CMD ["python", "main.py"]
\ No newline at end of file
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/06_eval/README.md b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/06_eval/README.md
new file mode 100644
index 0000000000..195d10e966
--- /dev/null
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/06_eval/README.md
@@ -0,0 +1,35 @@
+# Agent Framework Agent with Evaluation
+
+## Running the server locally
+
+Run the following command to start the server:
+
+```bash
+python main.py
+```
+
+## Interacting with the agent
+
+Send a POST request to the server with a JSON body containing a "message" field to interact with the agent. For example:
+
+```bash
+curl -X POST http://localhost:8088/responses -H "Content-Type: application/json" -d '{"input": "Hi"}'
+```
+
+The server will respond with a JSON object containing the response text and a response ID. You can use this response ID to continue the conversation in subsequent requests.
+
+## Multi-turn conversation
+
+To have a multi-turn conversation with the agent, include the previous response id in the request body. For example:
+
+```bash
+curl -X POST http://localhost:8088/responses -H "Content-Type: application/json" -d '{"input": "How are you?", "previous_response_id": "REPLACE_WITH_PREVIOUS_RESPONSE_ID"}'
+```
+
+## Deploying to Foundry
+
+TODO
+
+## Using the deployed agent in Agent Framework
+
+TODO
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/06_eval/agent.manifest.yaml b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/06_eval/agent.manifest.yaml
new file mode 100644
index 0000000000..a6cc8af7fd
--- /dev/null
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/06_eval/agent.manifest.yaml
@@ -0,0 +1,15 @@
+name: agent-framework-agent-with-eval
+description: >
+  An Agent Framework agent is evaluated on each response hosted by Foundry.
+metadata:
+  tags:
+    - AI Agent Hosting
+    - Azure AI AgentServer
+    - Responses Protocol
+    - Streaming
+template:
+  name: agent-framework-agent-with-eval
+  kind: hosted
+  protocols:
+    - protocol: responses
+      version: v0.1.0
\ No newline at end of file
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/06_eval/agent.yaml b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/06_eval/agent.yaml
new file mode 100644
index 0000000000..7fa752ca8b
--- /dev/null
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/06_eval/agent.yaml
@@ -0,0 +1,8 @@
+kind: hosted
+name: agent-framework-agent-with-eval
+protocols:
+  - protocol: responses
+    version: v0.1.0
+resources:
+  cpu: "0.25"
+  memory: 0.5Gi
\ No newline at end of file
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/06_eval/main.py b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/06_eval/main.py
new file mode 100644
index 0000000000..7c8b3f6dc8
--- /dev/null
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/06_eval/main.py
@@ -0,0 +1,50 @@
+# Copyright (c) Microsoft. All rights reserved.
+
+import os
+from random import randint
+
+from agent_framework import Agent, tool
+from agent_framework.foundry import FoundryChatClient
+from agent_framework_foundry_hosting import ResponsesHostServer
+from azure.ai.agentserver.responses import InMemoryResponseProvider
+from azure.identity import AzureCliCredential
+from dotenv import load_dotenv
+from pydantic import Field
+from typing_extensions import Annotated
+
+# Load environment variables from .env file
+load_dotenv()
+
+
+@tool(approval_mode="never_require")
+def get_weather(
+    location: Annotated[str, Field(description="The location to get the weather for.")],
+) -> str:
+    """Get the weather for a given location."""
+    conditions = ["sunny", "cloudy", "rainy", "stormy"]
+    return f"The weather in {location} is {conditions[randint(0, 3)]} with a high of {randint(10, 30)}°C."
+
+
+def main():
+    client = FoundryChatClient(
+        project_endpoint=os.environ["FOUNDRY_PROJECT_ENDPOINT"],
+        model=os.environ["FOUNDRY_MODEL"],
+        credential=AzureCliCredential(),
+    )
+
+    agent = Agent(
+        client=client,
+        instructions="You are a friendly assistant. Keep your answers brief.",
+        tools=[get_weather],
+        # History will be managed by the hosting infrastructure, thus there
+        # is no need to store history by the service. Learn more at:
+        # https://developers.openai.com/api/reference/resources/responses/methods/create
+        default_options={"store": False},
+    )
+
+    server = ResponsesHostServer(agent, provider=InMemoryResponseProvider())
+    server.run()
+
+
+if __name__ == "__main__":
+    main()
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/06_eval/requirements.txt b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/06_eval/requirements.txt
new file mode 100644
index 0000000000..f7dc62f3e3
--- /dev/null
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/06_eval/requirements.txt
@@ -0,0 +1,2 @@
+agent-framework
+agent-framework-foundry-hosting
\ No newline at end of file
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/07_foundry_memory/.dockerignore b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/07_foundry_memory/.dockerignore
new file mode 100644
index 0000000000..008e6e6616
--- /dev/null
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/07_foundry_memory/.dockerignore
@@ -0,0 +1,6 @@
+.venv
+__pycache__
+*.pyc
+*.pyo
+*.pyd
+.Python
\ No newline at end of file
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/07_foundry_memory/Dockerfile b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/07_foundry_memory/Dockerfile
new file mode 100644
index 0000000000..eaffb94f19
--- /dev/null
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/07_foundry_memory/Dockerfile
@@ -0,0 +1,16 @@
+FROM python:3.12-slim
+
+WORKDIR /app
+
+COPY . user_agent/
+WORKDIR /app/user_agent
+
+RUN if [ -f requirements.txt ]; then \
+        pip install -r requirements.txt; \
+    else \
+        echo "No requirements.txt found"; \
+    fi
+
+EXPOSE 8088
+
+CMD ["python", "main.py"]
\ No newline at end of file
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/07_foundry_memory/README.md b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/07_foundry_memory/README.md
new file mode 100644
index 0000000000..e9987de36d
--- /dev/null
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/07_foundry_memory/README.md
@@ -0,0 +1,35 @@
+# Agent Framework Agent with Foundry Memory
+
+## Running the server locally
+
+Run the following command to start the server:
+
+```bash
+python main.py
+```
+
+## Interacting with the agent
+
+Send a POST request to the server with a JSON body containing a "message" field to interact with the agent. For example:
+
+```bash
+curl -X POST http://localhost:8088/responses -H "Content-Type: application/json" -d '{"input": "Hi"}'
+```
+
+The server will respond with a JSON object containing the response text and a response ID. You can use this response ID to continue the conversation in subsequent requests.
+
+## Multi-turn conversation
+
+To have a multi-turn conversation with the agent, include the previous response id in the request body. For example:
+
+```bash
+curl -X POST http://localhost:8088/responses -H "Content-Type: application/json" -d '{"input": "How are you?", "previous_response_id": "REPLACE_WITH_PREVIOUS_RESPONSE_ID"}'
+```
+
+## Deploying to Foundry
+
+TODO
+
+## Using the deployed agent in Agent Framework
+
+TODO
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/07_foundry_memory/agent.manifest.yaml b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/07_foundry_memory/agent.manifest.yaml
new file mode 100644
index 0000000000..b0cf8c5fab
--- /dev/null
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/07_foundry_memory/agent.manifest.yaml
@@ -0,0 +1,15 @@
+name: agent-framework-agent-with-foundry-memory
+description: >
+  An Agent Framework agent with memory support hosted by Foundry.
+metadata:
+  tags:
+    - AI Agent Hosting
+    - Azure AI AgentServer
+    - Responses Protocol
+    - Streaming
+template:
+  name: agent-framework-agent-with-foundry-memory
+  kind: hosted
+  protocols:
+    - protocol: responses
+      version: v0.1.0
\ No newline at end of file
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/07_foundry_memory/agent.yaml b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/07_foundry_memory/agent.yaml
new file mode 100644
index 0000000000..90c9a26406
--- /dev/null
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/07_foundry_memory/agent.yaml
@@ -0,0 +1,8 @@
+kind: hosted
+name: agent-framework-agent-with-foundry-memory
+protocols:
+  - protocol: responses
+    version: v0.1.0
+resources:
+  cpu: "0.25"
+  memory: 0.5Gi
\ No newline at end of file
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/07_foundry_memory/main.py b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/07_foundry_memory/main.py
new file mode 100644
index 0000000000..1a46bdac3c
--- /dev/null
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/07_foundry_memory/main.py
@@ -0,0 +1,91 @@
+# Copyright (c) Microsoft. All rights reserved.
+
+import asyncio
+import logging
+import os
+from datetime import datetime, timezone
+
+from agent_framework import Agent
+from agent_framework.foundry import FoundryChatClient, FoundryMemoryProvider
+from agent_framework_foundry_hosting import ResponsesHostServer
+from azure.ai.agentserver.responses import InMemoryResponseProvider
+from azure.ai.projects.aio import AIProjectClient
+from azure.ai.projects.models import (
+    MemoryStoreDefaultDefinition,
+    MemoryStoreDefaultOptions,
+)
+from azure.identity import AzureCliCredential
+from dotenv import load_dotenv
+
+# Load environment variables from .env file
+load_dotenv()
+
+logging.basicConfig(level=logging.INFO)
+
+
+async def _create_memory_store(project_client: AIProjectClient) -> FoundryMemoryProvider:
+    memory_store_name = f"hosted_agent_memory_{datetime.now(timezone.utc).strftime('%Y%m%d')}"
+    options = MemoryStoreDefaultOptions(
+        chat_summary_enabled=True,
+        user_profile_enabled=True,
+        user_profile_details=(
+            "Avoid irrelevant or sensitive data, such as age, financials, precise location, and credentials"
+        ),
+    )
+    memory_store_definition = MemoryStoreDefaultDefinition(
+        chat_model=os.environ["FOUNDRY_MODEL"],
+        embedding_model=os.environ["AZURE_OPENAI_EMBEDDING_MODEL"],
+        options=options,
+    )
+    memory_store = await project_client.beta.memory_stores.create(
+        name=memory_store_name,
+        description="Memory store for Agent Framework with FoundryMemoryProvider",
+        definition=memory_store_definition,
+    )
+
+    return FoundryMemoryProvider(
+        project_client=project_client,
+        memory_store_name=memory_store.name,
+        # Scope memories to a specific user, if not set, the session_id
+        # will be used as scope, which means memories are only shared within the same session
+        scope="demo",
+        # Do not wait to update memories after each interaction (for demo purposes)
+        # In production, consider setting a delay to batch updates and reduce costs
+        update_delay=0,
+    )
+
+
+async def _delete_memory_store(project_client: AIProjectClient, memory_store_name: str):
+    await project_client.beta.memory_stores.delete(name=memory_store_name)
+
+
+async def main():
+    client = FoundryChatClient(
+        project_endpoint=os.environ["FOUNDRY_PROJECT_ENDPOINT"],
+        model=os.environ["FOUNDRY_MODEL"],
+        credential=AzureCliCredential(),
+    )
+
+    # Create the memory store
+    memory_provider = await _create_memory_store(client.project_client)
+
+    agent = Agent(
+        client=client,
+        instructions="You are a friendly assistant. Keep your answers brief.",
+        context_providers=[memory_provider],
+        # History will be managed by the hosting infrastructure, thus there
+        # is no need to store history by the service. Learn more at:
+        # https://developers.openai.com/api/reference/resources/responses/methods/create
+        default_options={"store": False},
+    )
+
+    server = ResponsesHostServer(agent, provider=InMemoryResponseProvider())
+
+    try:
+        await server.run_async()
+    finally:
+        await _delete_memory_store(client.project_client, memory_provider.memory_store_name)
+
+
+if __name__ == "__main__":
+    asyncio.run(main())
diff --git a/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/07_foundry_memory/requirements.txt b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/07_foundry_memory/requirements.txt
new file mode 100644
index 0000000000..f7dc62f3e3
--- /dev/null
+++ b/python/samples/05-end-to-end/hosted_agents/foundry_hosted_responses/07_foundry_memory/requirements.txt
@@ -0,0 +1,2 @@
+agent-framework
+agent-framework-foundry-hosting
\ No newline at end of file

From 4d7c035bb335b74aee28f6700f83cc04af8ecf6b Mon Sep 17 00:00:00 2001
From: Tao Chen <taochen@microsoft.com>
Date: Fri, 10 Apr 2026 09:52:50 -0700
Subject: [PATCH 12/13] Update user agent prefix

---
 .../agent_framework_foundry_hosting/_invocations.py             | 2 +-
 .../agent_framework_foundry_hosting/_responses.py               | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/python/packages/foundry_hosting/agent_framework_foundry_hosting/_invocations.py b/python/packages/foundry_hosting/agent_framework_foundry_hosting/_invocations.py
index 2feb1001ba..08407874df 100644
--- a/python/packages/foundry_hosting/agent_framework_foundry_hosting/_invocations.py
+++ b/python/packages/foundry_hosting/agent_framework_foundry_hosting/_invocations.py
@@ -11,7 +11,7 @@
 class InvocationsHostServer(InvocationAgentServerHost):
     """An invocations server host for an agent."""
 
-    USER_AGENT_PREFIX = "foundry-hosting-invocations"
+    USER_AGENT_PREFIX = "foundry-hosting"
 
     def __init__(
         self,
diff --git a/python/packages/foundry_hosting/agent_framework_foundry_hosting/_responses.py b/python/packages/foundry_hosting/agent_framework_foundry_hosting/_responses.py
index 8c0447d7bd..94c710f5b8 100644
--- a/python/packages/foundry_hosting/agent_framework_foundry_hosting/_responses.py
+++ b/python/packages/foundry_hosting/agent_framework_foundry_hosting/_responses.py
@@ -41,7 +41,7 @@
 class ResponsesHostServer(ResponsesAgentServerHost):
     """A responses server host for an agent."""
 
-    USER_AGENT_PREFIX = "foundry-hosting-responses"
+    USER_AGENT_PREFIX = "foundry-hosting"
 
     def __init__(
         self,

From 6f85956fd9cfbc3fdec2a0734b91108a083ce9cb Mon Sep 17 00:00:00 2001
From: Tao Chen <taochen@microsoft.com>
Date: Fri, 10 Apr 2026 10:00:16 -0700
Subject: [PATCH 13/13] Update user agent prefix doc

---
 python/packages/core/agent_framework/_telemetry.py | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/python/packages/core/agent_framework/_telemetry.py b/python/packages/core/agent_framework/_telemetry.py
index f4d259806d..8e59704b8a 100644
--- a/python/packages/core/agent_framework/_telemetry.py
+++ b/python/packages/core/agent_framework/_telemetry.py
@@ -36,7 +36,7 @@ def append_to_user_agent(prefix: str) -> None:
     Duplicate prefixes are ignored.
 
     Args:
-        prefix: The prefix to prepend (e.g. "foundry-hosting-responses").
+        prefix: The prefix to prepend (e.g. "foundry-hosting").
     """
     if prefix and prefix not in _user_agent_prefixes:
         _user_agent_prefixes.append(prefix)