chore: import upstream snapshot with attribution

python/packages/gemini/AGENTS.md

@@ -0,0 +1,36 @@
+# Gemini Package (agent-framework-gemini)
+
+Integration with Google's Gemini Developer API and Vertex AI via the `google-genai` SDK.
+
+## Core Classes
+
+- **`RawGeminiChatClient`** - Lightweight chat client without any layers, for custom pipeline composition
+- **`GeminiChatClient`** - Full-featured chat client with function invocation, middleware, and telemetry
+- **`GeminiChatOptions`** - Options TypedDict for Gemini-specific parameters
+- **`GeminiSettings`** - Settings loaded from environment variables
+- **`GoogleGeminiSettings`** - SDK-standard `GOOGLE_*` settings loaded from environment variables
+- **`ThinkingConfig`** - Configuration for extended thinking
+
+## Gemini-specific Options
+
+- **`thinking_config`** - Enable extended thinking via `ThinkingConfig`
+- **`response_schema`** - Raw JSON schema dict for structured output (alternative to `response_format`)
+- **`top_k`** - Top-K sampling parameter
+
+## Built-in Tool Factory Methods
+
+- **`get_web_search_tool()`** - Google Search grounding for up-to-date web answers
+- **`get_code_interpreter_tool()`** - Sandboxed code execution
+- **`get_maps_grounding_tool()`** - Google Maps grounding for location and mapping
+- **`get_file_search_tool()`** - Retrieval from Gemini file search stores
+- **`get_mcp_tool()`** - Model Context Protocol server integration
+
+## Usage
+
+```python
+from agent_framework import Content, Message
+from agent_framework_gemini import GeminiChatClient
+
+client = GeminiChatClient(model="gemini-2.5-flash")
+response = await client.get_response([Message(role="user", contents=[Content.from_text("Hello")])])
+```

python/packages/gemini/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

python/packages/gemini/README.md

@@ -0,0 +1,51 @@
+# Get Started with Microsoft Agent Framework Gemini
+
+Install the provider package:
+
+```bash
+pip install agent-framework-gemini --pre
+```
+
+## Gemini Integration
+
+The Gemini integration enables Microsoft Agent Framework applications to call Google Gemini models with familiar chat abstractions, including streaming, tool/function calling, and structured output.
+
+## Structured Output
+
+Gemini structured output can be configured with either a Pydantic model in `response_format`, a JSON schema mapping in `response_format`, or a Gemini-specific `response_schema`. Declarative agents that define `outputSchema` pass that schema through `response_format`.
+
+## Authentication
+
+The connector supports both `google-genai` authentication modes.
+
+### Gemini Developer API
+
+Obtain an API key from [Google AI Studio](https://aistudio.google.com/apikey) and set either the package-prefixed or SDK-standard environment variable:
+
+```bash
+export GEMINI_API_KEY="your-api-key"
+# or: export GOOGLE_API_KEY="your-api-key"
+export GEMINI_MODEL="gemini-2.5-flash-lite"
+# or: export GOOGLE_MODEL="gemini-2.5-flash-lite"
+```
+
+### Vertex AI
+
+Set the standard Vertex AI environment variables used by `google-genai`:
+
+```bash
+export GOOGLE_GENAI_USE_VERTEXAI=true
+export GOOGLE_CLOUD_PROJECT="your-project-id"
+export GOOGLE_CLOUD_LOCATION="global"
+export GOOGLE_MODEL="gemini-2.5-flash-lite"
+```
+
+## Examples
+
+See the [Google Gemini samples](samples/) for runnable end-to-end scripts covering:
+
+- Basic agent with tool calling and streaming
+- Extended thinking with `ThinkingConfig`
+- Google Search grounding
+- Google Maps grounding
+- Built-in code execution

python/packages/gemini/agent_framework_gemini/__init__.py

@@ -0,0 +1,27 @@
+# Copyright (c) Microsoft. All rights reserved.
+
+import importlib.metadata
+
+from ._chat_client import (
+    GeminiChatClient,
+    GeminiChatOptions,
+    GeminiSettings,
+    GoogleGeminiSettings,
+    RawGeminiChatClient,
+    ThinkingConfig,
+)
+
+try:
+    __version__ = importlib.metadata.version(__name__)
+except importlib.metadata.PackageNotFoundError:
+    __version__ = "0.0.0"
+
+__all__ = [
+    "GeminiChatClient",
+    "GeminiChatOptions",
+    "GeminiSettings",
+    "GoogleGeminiSettings",
+    "RawGeminiChatClient",
+    "ThinkingConfig",
+    "__version__",
+]

python/packages/gemini/pyproject.toml

@@ -0,0 +1,105 @@
+[project]
+name = "agent-framework-gemini"
+description = "Google Gemini integration for Microsoft Agent Framework."
+authors = [{ name = "Microsoft", email = "af-support@microsoft.com"}]
+readme = "README.md"
+requires-python = ">=3.10"
+version = "1.0.0a260709"
+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 :: 3 - Alpha",
+  "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",
+  "Framework :: Pydantic :: 2",
+  "Typing :: Typed",
+]
+dependencies = [
+    "agent-framework-core>=1.11.0,<2",
+    "google-genai>=1.68.0,<3.0.0",
+]
+
+[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 = []
+markers = [
+    "integration: marks tests as integration tests that require external services",
+    "flaky: marks tests as flaky and eligible for automatic retry",
+]
+timeout = 120
+
+[tool.ruff]
+extend = "../../pyproject.toml"
+
+[tool.ruff.lint.extend-per-file-ignores]
+"samples/**" = ["S", "T201"]
+
+[tool.coverage.run]
+omit = [
+    "**/__init__.py"
+]
+
+[tool.pyright]
+extends = "../../pyproject.toml"
+include = ["agent_framework_gemini"]
+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_gemini"]
+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_gemini"
+
+[tool.poe.tasks.test]
+help = "Run the default unit test suite for this package."
+cmd = 'pytest -m "not integration" --cov=agent_framework_gemini --cov-report=term-missing:skip-covered tests'
+
+[tool.flit.module]
+name = "agent_framework_gemini"
+
+[build-system]
+requires = ["flit-core >= 3.11,<4.0"]
+build-backend = "flit_core.buildapi"

python/packages/gemini/samples/README.md

@@ -0,0 +1,20 @@
+# Google Gemini Examples
+
+This folder contains examples demonstrating how to use Google Gemini models with the Agent Framework.
+
+## Examples
+
+| File | Description |
+|------|-------------|
+| [`gemini_basic.py`](gemini_basic.py) | Basic agent with a weather tool, demonstrating both streaming and non-streaming responses. |
+| [`gemini_advanced.py`](gemini_advanced.py) | Extended thinking via `ThinkingConfig` for reasoning-heavy questions (Gemini 2.5+). |
+| [`gemini_with_google_search.py`](gemini_with_google_search.py) | Google Search grounding for up-to-date answers. |
+| [`gemini_with_google_maps.py`](gemini_with_google_maps.py) | Google Maps grounding for location and mapping information. |
+| [`gemini_with_code_execution.py`](gemini_with_code_execution.py) | Built-in code execution tool for computing precise answers in a sandboxed environment. |
+
+## Environment Variables
+
+- `GOOGLE_MODEL` or `GEMINI_MODEL`: The Gemini model to use (for example,
+  `gemini-2.5-flash-lite` or `gemini-2.5-pro`)
+- For Gemini Developer API: `GEMINI_API_KEY` or `GOOGLE_API_KEY`
+- For Vertex AI: `GOOGLE_GENAI_USE_VERTEXAI=true`, `GOOGLE_CLOUD_PROJECT`, and `GOOGLE_CLOUD_LOCATION`

python/packages/gemini/samples/__init__.py

@@ -0,0 +1 @@
+# Copyright (c) Microsoft. All rights reserved.

python/packages/gemini/samples/gemini_advanced.py

@@ -0,0 +1,59 @@
+# Copyright (c) Microsoft. All rights reserved.
+
+"""Shows how to enable extended thinking with ThinkingConfig.
+
+Allows the model to reason through complex problems before responding.
+
+Requires ``GOOGLE_MODEL`` or ``GEMINI_MODEL`` and either Gemini Developer API credentials
+(``GEMINI_API_KEY`` or ``GOOGLE_API_KEY``) or Vertex AI settings
+(``GOOGLE_GENAI_USE_VERTEXAI``, ``GOOGLE_CLOUD_PROJECT``, and ``GOOGLE_CLOUD_LOCATION``).
+"""
+
+import asyncio
+
+from agent_framework import Agent
+from dotenv import load_dotenv
+
+from agent_framework_gemini import GeminiChatClient, GeminiChatOptions, ThinkingConfig
+
+load_dotenv()
+
+
+async def main() -> None:
+    """Example of extended thinking with a Python version comparison question."""
+    print("=== Extended thinking ===")
+
+    # 1. Configure Gemini extended thinking for a reasoning-heavy request.
+    options: GeminiChatOptions = {
+        "thinking_config": ThinkingConfig(thinking_budget=2048),
+    }
+
+    # 2. Create the agent with the Gemini chat client and default thinking options.
+    agent = Agent(
+        client=GeminiChatClient(),
+        name="PythonAgent",
+        instructions="You are a helpful Python expert.",
+        default_options=options,
+    )
+
+    # 3. Stream the answer so you can see the final response as it arrives.
+    query = "What new language features were introduced in Python between 3.10 and 3.14?"
+    print(f"User: {query}")
+    print("Agent: ", end="", flush=True)
+    async for chunk in agent.run(query, stream=True):
+        if chunk.text:
+            print(chunk.text, end="", flush=True)
+    print("\n")
+
+
+if __name__ == "__main__":
+    asyncio.run(main())
+
+"""
+Sample output:
+=== Extended thinking ===
+User: What new language features were introduced in Python between 3.10 and 3.14?
+Agent: Python 3.11 introduced exception groups and TaskGroup.
+Python 3.12 added PEP 695 type parameter syntax.
+Python 3.13-3.14 continued improving typing, performance, and developer ergonomics.
+"""

python/packages/gemini/samples/gemini_basic.py

@@ -0,0 +1,93 @@
+# Copyright (c) Microsoft. All rights reserved.
+
+"""Shows how to use GeminiChatClient with an agent and a custom tool.
+
+Covers both non-streaming and streaming responses.
+
+Requires ``GOOGLE_MODEL`` or ``GEMINI_MODEL`` and either Gemini Developer API credentials
+(``GEMINI_API_KEY`` or ``GOOGLE_API_KEY``) or Vertex AI settings
+(``GOOGLE_GENAI_USE_VERTEXAI``, ``GOOGLE_CLOUD_PROJECT``, and ``GOOGLE_CLOUD_LOCATION``).
+"""
+
+import asyncio
+from random import randint
+from typing import Annotated
+
+from agent_framework import Agent, tool
+from dotenv import load_dotenv
+
+from agent_framework_gemini import GeminiChatClient
+
+load_dotenv()
+
+
+# NOTE: approval_mode="never_require" is for sample brevity. Use "always_require" in production
+@tool(approval_mode="never_require")
+def get_weather(
+    location: Annotated[str, "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."
+
+
+async def non_streaming_example() -> None:
+    """Runs the agent and waits for the complete response before printing it."""
+    print("=== Non-streaming ===")
+
+    # 1. Create the agent with the Gemini chat client and local weather tool.
+    agent = Agent(
+        client=GeminiChatClient(),
+        name="WeatherAgent",
+        instructions="You are a helpful weather agent.",
+        tools=[get_weather],
+    )
+
+    # 2. Ask the agent for a single weather lookup and print the final response.
+    query = "What's the weather like in Karlsruhe, Germany?"
+    print(f"User: {query}")
+    result = await agent.run(query)
+    print(f"Result: {result}\n")
+
+
+async def streaming_example() -> None:
+    """Runs the agent and prints each chunk as it is received."""
+    print("=== Streaming ===")
+
+    # 1. Create the same agent configuration for a streaming tool-call example.
+    agent = Agent(
+        client=GeminiChatClient(),
+        name="WeatherAgent",
+        instructions="You are a helpful weather agent.",
+        tools=[get_weather],
+    )
+
+    # 2. Ask a multi-location question and stream the model output as it arrives.
+    query = "What's the weather like in Portland and in Paris?"
+    print(f"User: {query}")
+    print("Agent: ", end="", flush=True)
+    async for chunk in agent.run(query, stream=True):
+        if chunk.text:
+            print(chunk.text, end="", flush=True)
+    print("\n")
+
+
+async def main() -> None:
+    """Run non-streaming and streaming examples."""
+    await non_streaming_example()
+    await streaming_example()
+
+
+if __name__ == "__main__":
+    asyncio.run(main())
+
+"""
+Sample output:
+=== Non-streaming ===
+User: What's the weather like in Karlsruhe, Germany?
+Result: The weather in Karlsruhe, Germany is currently sunny with a high of 16°C.
+
+=== Streaming ===
+User: What's the weather like in Portland and in Paris?
+Agent: In Portland, it is currently rainy with a high of 11°C. In Paris, it is cloudy with a high of 27°C.
+"""

python/packages/gemini/samples/gemini_with_code_execution.py

@@ -0,0 +1,52 @@
+# Copyright (c) Microsoft. All rights reserved.
+
+"""Shows how to enable Gemini's built-in code execution tool.
+
+Allows the model to write and run code in a sandboxed environment to answer questions.
+
+Requires ``GOOGLE_MODEL`` or ``GEMINI_MODEL`` and either Gemini Developer API credentials
+(``GEMINI_API_KEY`` or ``GOOGLE_API_KEY``) or Vertex AI settings
+(``GOOGLE_GENAI_USE_VERTEXAI``, ``GOOGLE_CLOUD_PROJECT``, and ``GOOGLE_CLOUD_LOCATION``).
+"""
+
+import asyncio
+
+from agent_framework import Agent
+from dotenv import load_dotenv
+
+from agent_framework_gemini import GeminiChatClient
+
+load_dotenv()
+
+
+async def main() -> None:
+    """Run the code execution example."""
+    print("=== Code execution ===")
+
+    # 1. Create the agent with Gemini and the built-in code execution tool.
+    agent = Agent(
+        client=GeminiChatClient(),
+        name="CodeAgent",
+        instructions="You are a helpful assistant. Use code execution to compute precise answers.",
+        tools=[GeminiChatClient.get_code_interpreter_tool()],
+    )
+
+    # 2. Ask for a computed answer and stream the generated code and final result.
+    query = "What are the first 20 prime numbers? Compute them in code."
+    print(f"User: {query}")
+    print("Agent: ", end="", flush=True)
+    async for chunk in agent.run(query, stream=True):
+        if chunk.text:
+            print(chunk.text, end="", flush=True)
+    print("\n")
+
+
+if __name__ == "__main__":
+    asyncio.run(main())
+
+"""
+Sample output:
+=== Code execution ===
+User: What are the first 20 prime numbers? Compute them in code.
+Agent: The first 20 prime numbers are 2, 3, 5, 7, 11, 13, 17, 19, 23, 29, 31, 37, 41, 43, 47, 53, 59, 61, 67, and 71.
+"""

python/packages/gemini/samples/gemini_with_google_maps.py

@@ -0,0 +1,53 @@
+# Copyright (c) Microsoft. All rights reserved.
+
+"""Shows how to enable Google Maps grounding.
+
+Allows Gemini to retrieve location and mapping information before responding.
+
+Requires ``GOOGLE_MODEL`` or ``GEMINI_MODEL`` and either Gemini Developer API credentials
+(``GEMINI_API_KEY`` or ``GOOGLE_API_KEY``) or Vertex AI settings
+(``GOOGLE_GENAI_USE_VERTEXAI``, ``GOOGLE_CLOUD_PROJECT``, and ``GOOGLE_CLOUD_LOCATION``).
+"""
+
+import asyncio
+
+from agent_framework import Agent
+from dotenv import load_dotenv
+
+from agent_framework_gemini import GeminiChatClient
+
+load_dotenv()
+
+
+async def main() -> None:
+    """Run the Google Maps grounding example."""
+    print("=== Google Maps grounding ===")
+
+    # 1. Create the agent with Gemini and the built-in Google Maps grounding tool.
+    agent = Agent(
+        client=GeminiChatClient(),
+        name="MapsAgent",
+        instructions="You are a helpful travel assistant. Use Google Maps to provide accurate location information.",
+        tools=[GeminiChatClient.get_maps_grounding_tool()],
+    )
+
+    # 2. Ask a location-aware question and stream the grounded answer.
+    query = "What are some highly rated restaurants in the city center of Karlsruhe, Germany?"
+    print(f"User: {query}")
+    print("Agent: ", end="", flush=True)
+    async for chunk in agent.run(query, stream=True):
+        if chunk.text:
+            print(chunk.text, end="", flush=True)
+    print("\n")
+
+
+if __name__ == "__main__":
+    asyncio.run(main())
+
+"""
+Sample output:
+=== Google Maps grounding ===
+User: What are some highly rated restaurants in the city center of Karlsruhe, Germany?
+Agent: Here are several highly rated restaurants near Karlsruhe city center,
+along with their cuisine styles and approximate walking distance.
+"""

python/packages/gemini/samples/gemini_with_google_search.py

@@ -0,0 +1,52 @@
+# Copyright (c) Microsoft. All rights reserved.
+
+"""Shows how to enable Google Search grounding.
+
+Allows Gemini to retrieve up-to-date information from the web before responding.
+
+Requires ``GOOGLE_MODEL`` or ``GEMINI_MODEL`` and either Gemini Developer API credentials
+(``GEMINI_API_KEY`` or ``GOOGLE_API_KEY``) or Vertex AI settings
+(``GOOGLE_GENAI_USE_VERTEXAI``, ``GOOGLE_CLOUD_PROJECT``, and ``GOOGLE_CLOUD_LOCATION``).
+"""
+
+import asyncio
+
+from agent_framework import Agent
+from dotenv import load_dotenv
+
+from agent_framework_gemini import GeminiChatClient
+
+load_dotenv()
+
+
+async def main() -> None:
+    """Run the Google Search grounding example."""
+    print("=== Google Search grounding ===")
+
+    # 1. Create the agent with Gemini and the built-in Google Search grounding tool.
+    agent = Agent(
+        client=GeminiChatClient(),
+        name="SearchAgent",
+        instructions="You are a helpful assistant. Use Google Search to provide accurate, up-to-date answers.",
+        tools=[GeminiChatClient.get_web_search_tool()],
+    )
+
+    # 2. Ask a current-events style question and stream the grounded answer.
+    query = "What is the latest stable release of the .NET SDK?"
+    print(f"User: {query}")
+    print("Agent: ", end="", flush=True)
+    async for chunk in agent.run(query, stream=True):
+        if chunk.text:
+            print(chunk.text, end="", flush=True)
+    print("\n")
+
+
+if __name__ == "__main__":
+    asyncio.run(main())
+
+"""
+Sample output:
+=== Google Search grounding ===
+User: What is the latest stable release of the .NET SDK?
+Agent: As of April 14, 2026, the latest stable release of the .NET SDK is .NET 10.0 (SDK 10.0.201).
+"""