エージェントへのミドルウェアの追加

[前提条件]

前提条件と NuGet パッケージのインストールについては、このチュートリアルの 「エージェントの作成と実行 」の手順を参照してください。

手順 1: 単純なエージェントを作成する

まず、関数ツールを使用して基本的なエージェントを作成します。

using System;
using System.ComponentModel;
using Azure.AI.Projects;
using Azure.Identity;
using Microsoft.Agents.AI;
using Microsoft.Extensions.AI;

[Description("The current datetime offset.")]
static string GetDateTime()
    => DateTimeOffset.Now.ToString();

AIAgent baseAgent = new AIProjectClient(
    new Uri("<your-foundry-project-endpoint>"),
    new DefaultAzureCredential())
        .AsAIAgent(
            model: "gpt-4o-mini",
            instructions: "You are an AI assistant that helps people find information.",
            tools: [AIFunctionFactory.Create(GetDateTime, name: nameof(GetDateTime))]);

手順 2: エージェント実行ミドルウェアを作成する

次に、エージェントの実行ごとに呼び出される関数を作成します。 エージェントからの入力と出力を検査できます。

ミドルウェアを使用して実行を停止する意図がない限り、関数は指定されたRunAsyncでinnerAgentを呼び出す必要があります。

このサンプル ミドルウェアは、エージェントの実行からの入力と出力を検査し、エージェントとの間で送受信されるメッセージの数を出力します。

using System.Collections.Generic;
using System.Linq;
using System.Threading;
using System.Threading.Tasks;

async Task<AgentResponse> CustomAgentRunMiddleware(
    IEnumerable<ChatMessage> messages,
    AgentSession? session,
    AgentRunOptions? options,
    AIAgent innerAgent,
    CancellationToken cancellationToken)
{
    Console.WriteLine($"Input: {messages.Count()}");
    var response = await innerAgent.RunAsync(messages, session, options, cancellationToken).ConfigureAwait(false);
    Console.WriteLine($"Output: {response.Messages.Count}");
    return response;
}

手順 3: エージェント実行ミドルウェアをエージェントに追加する

手順 1 で作成した baseAgent にこのミドルウェア関数を追加するには、ビルダー パターンを使用します。 これにより、ミドルウェアが適用された新しいエージェントが作成されます。 元の baseAgent は変更されません。

var middlewareEnabledAgent = baseAgent
    .AsBuilder()
        .Use(runFunc: CustomAgentRunMiddleware, runStreamingFunc: null)
    .Build();

クエリを使用してエージェントを実行すると、ミドルウェアが呼び出され、入力メッセージの数と応答メッセージの数が出力されます。

Console.WriteLine(await middlewareEnabledAgent.RunAsync("What's the current time?"));

手順 4: 関数呼び出しミドルウェアを作成する

また、呼び出される関数ツールごとに呼び出されるミドルウェアを作成することもできます。 呼び出される関数と関数呼び出しの結果を検査または変更できる関数呼び出しミドルウェアの例を次に示します。

ミドルウェアを使用して関数ツールを実行しない場合を除き、ミドルウェアは提供された nextFuncを呼び出す必要があります。

using System.Threading;
using System.Threading.Tasks;

async ValueTask<object?> CustomFunctionCallingMiddleware(
    AIAgent agent,
    FunctionInvocationContext context,
    Func<FunctionInvocationContext, CancellationToken, ValueTask<object?>> next,
    CancellationToken cancellationToken)
{
    Console.WriteLine($"Function Name: {context!.Function.Name}");
    var result = await next(context, cancellationToken);
    Console.WriteLine($"Function Call Result: {result}");

    return result;
}

手順 5: 関数呼び出しミドルウェアをエージェントに追加する

エージェント実行ミドルウェアの追加と同じように、次のように関数呼び出しミドルウェアを追加できます。

var middlewareEnabledAgent = baseAgent
    .AsBuilder()
        .Use(CustomFunctionCallingMiddleware)
    .Build();

関数を呼び出すクエリを使用してエージェントを実行すると、ミドルウェアが呼び出され、関数名と呼び出し結果が出力されます。

Console.WriteLine(await middlewareEnabledAgent.RunAsync("What's the current time?"));

手順 6: チャット クライアント ミドルウェアを作成する

IChatClientを使用して構築されたエージェントの場合は、エージェントからIChatClientへの呼び出しを傍受したいかもしれません。 この場合、 IChatClientにミドルウェアを使用できます。

チャット クライアントが提供する推論サービスへの要求の入力と出力を検査または変更できるチャット クライアント ミドルウェアの例を次に示します。

using System.Collections.Generic;
using System.Linq;
using System.Threading;
using System.Threading.Tasks;

async Task<ChatResponse> CustomChatClientMiddleware(
    IEnumerable<ChatMessage> messages,
    ChatOptions? options,
    IChatClient innerChatClient,
    CancellationToken cancellationToken)
{
    Console.WriteLine($"Input: {messages.Count()}");
    var response = await innerChatClient.GetResponseAsync(messages, options, cancellationToken);
    Console.WriteLine($"Output: {response.Messages.Count}");

    return response;
}

手順 7: チャットクライアントミドルウェアをIChatClientに追加します。

ミドルウェアを IChatClientに追加するには、ビルダー パターンを使用できます。 ミドルウェアを追加した後は、通常どおりエージェントで IChatClient を使用できます。

var chatClient = new AIProjectClient(
    new Uri("<your-foundry-project-endpoint>"),
    new DefaultAzureCredential())
        .GetProjectOpenAIClient()
        .GetProjectResponsesClient()
        .AsIChatClient("gpt-4o-mini");

var middlewareEnabledChatClient = chatClient
    .AsBuilder()
        .Use(getResponseFunc: CustomChatClientMiddleware, getStreamingResponseFunc: null)
    .Build();

var agent = new ChatClientAgent(middlewareEnabledChatClient, instructions: "You are a helpful assistant.");

IChatClient SDK クライアントのヘルパー メソッドの 1 つを使用してエージェントを構築するときに、ファクトリ メソッドを使用してミドルウェアを登録することもできます。

var agent = new AIProjectClient(
    new Uri("<your-foundry-project-endpoint>"),
    new DefaultAzureCredential())
        .AsAIAgent(
            model: "gpt-4o-mini",
            instructions: "You are a helpful assistant.",
            clientFactory: (chatClient) => chatClient
                .AsBuilder()
                    .Use(getResponseFunc: CustomChatClientMiddleware, getStreamingResponseFunc: null)
                .Build());

手順 1: 単純なエージェントを作成する

まず、基本的なエージェントを作成します。

import asyncio
from agent_framework import Agent
from agent_framework.foundry import FoundryChatClient
from azure.identity.aio import AzureCliCredential

async def main():
    credential = AzureCliCredential()

    async with Agent(

        client=FoundryChatClient(credential=credential),
        name="GreetingAgent",
        instructions="You are a friendly greeting assistant.",
    ) as agent:
        result = await agent.run("Hello!")
        print(result.text)

if __name__ == "__main__":
    asyncio.run(main())

手順 2: ミドルウェアを作成する

エージェントがいつ実行されているかを確認する単純なログ ミドルウェアを作成します。

from collections.abc import Awaitable, Callable

from agent_framework import AgentContext

async def logging_agent_middleware(
    context: AgentContext,
    call_next: Callable[[], Awaitable[None]],
) -> None:
    """Simple middleware that logs agent execution."""
    print("Agent starting...")

    # Continue to agent execution
    await call_next()

    print("Agent finished!")

手順 3: エージェントにミドルウェアを追加する

エージェントの作成時にミドルウェアを追加します。

async def main():
    credential = AzureCliCredential()

    async with Agent(

        client=FoundryChatClient(credential=credential),
        name="GreetingAgent",
        instructions="You are a friendly greeting assistant.",
        middleware=[logging_agent_middleware],  # Add your middleware here
    ) as agent:
        result = await agent.run("Hello!")
        print(result.text)

手順 4: 関数ミドルウェアを作成する

エージェントで関数を使用している場合は、ツールを実行する前に関数呼び出しをインターセプトし、ツールのみのランタイム値を設定できます。

from collections.abc import Awaitable, Callable

from agent_framework import FunctionInvocationContext

def get_time(ctx: FunctionInvocationContext) -> str:
    """Get the current time."""
    from datetime import datetime
    source = ctx.kwargs.get("request_source", "direct")
    return f"[{source}] {datetime.now().strftime('%H:%M:%S')}"

async def inject_function_kwargs(
    context: FunctionInvocationContext,
    call_next: Callable[[], Awaitable[None]],
) -> None:
    """Middleware that adds tool-only runtime values before execution."""
    context.kwargs.setdefault("request_source", "middleware")

    await call_next()

# Add both the function and middleware to your agent
async with Agent(
    client=FoundryChatClient(credential=credential),
    name="TimeAgent",
    instructions="You can tell the current time.",
    tools=[get_time],
    middleware=[inject_function_kwargs],
) as agent:
    result = await agent.run("What time is it?")

手順 5: Run-Level ミドルウェアを使用する

特定の実行のミドルウェアを追加することもできます。

# Use middleware for this specific run only
result = await agent.run(
    "This is important!",
    middleware=[logging_function_middleware]
)

次は何ですか？

より高度なシナリオについては、次の エージェント ミドルウェア ユーザー ガイドを参照してください。

• さまざまな種類のミドルウェア (エージェント、関数、チャット)。
• 複雑なシナリオ用のクラス ベースのミドルウェア。
• ミドルウェアの終了と結果のオーバーライド。
• 高度なミドルウェア パターンとベスト プラクティス。

完全なサンプル例

クラス ベースのミドルウェア

# Copyright (c) Microsoft. All rights reserved.

import asyncio
import time
from collections.abc import Awaitable, Callable
from random import randint
from typing import Annotated

from agent_framework import (
    AgentContext,
    AgentMiddleware,
    AgentResponse,
    FunctionInvocationContext,
    FunctionMiddleware,
    Message,
    tool,
)
from agent_framework import Agent
from agent_framework.foundry import FoundryChatClient
from azure.identity.aio import AzureCliCredential
from pydantic import Field

"""
Class-based MiddlewareTypes Example

This sample demonstrates how to implement middleware using class-based approach by inheriting
from AgentMiddleware and FunctionMiddleware base classes. The example includes:

- SecurityAgentMiddleware: Checks for security violations in user queries and blocks requests
  containing sensitive information like passwords or secrets
- LoggingFunctionMiddleware: Logs function execution details including timing and parameters

This approach is useful when you need stateful middleware or complex logic that benefits
from object-oriented design patterns.
"""


# NOTE: approval_mode="never_require" is for sample brevity. Use "always_require" in production; see samples/02-agents/tools/function_tool_with_approval.py and samples/02-agents/tools/function_tool_with_approval_and_sessions.py.
@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."


class SecurityAgentMiddleware(AgentMiddleware):
    """Agent middleware that checks for security violations."""

    async def process(
        self,
        context: AgentContext,
        call_next: Callable[[], Awaitable[None]],
    ) -> None:
        # Check for potential security violations in the query
        # Look at the last user message
        last_message = context.messages[-1] if context.messages else None
        if last_message and last_message.text:
            query = last_message.text
            if "password" in query.lower() or "secret" in query.lower():
                print("[SecurityAgentMiddleware] Security Warning: Detected sensitive information, blocking request.")
                # Override the result with warning message
                context.result = AgentResponse(
                    messages=[Message("assistant", ["Detected sensitive information, the request is blocked."])]
                )
                # Simply don't call call_next() to prevent execution
                return

        print("[SecurityAgentMiddleware] Security check passed.")
        await call_next()


class LoggingFunctionMiddleware(FunctionMiddleware):
    """Function middleware that logs function calls."""

    async def process(
        self,
        context: FunctionInvocationContext,
        call_next: Callable[[], Awaitable[None]],
    ) -> None:
        function_name = context.function.name
        print(f"[LoggingFunctionMiddleware] About to call function: {function_name}.")

        start_time = time.time()

        await call_next()

        end_time = time.time()
        duration = end_time - start_time

        print(f"[LoggingFunctionMiddleware] Function {function_name} completed in {duration:.5f}s.")


async def main() -> None:
    """Example demonstrating class-based middleware."""
    print("=== Class-based MiddlewareTypes Example ===")

    # For authentication, run `az login` command in terminal or replace AzureCliCredential with preferred
    # authentication option.
    async with (
        AzureCliCredential() as credential,
        Agent(
            client=FoundryChatClient(credential=credential),
            name="WeatherAgent",
            instructions="You are a helpful weather assistant.",
            tools=get_weather,
            middleware=[SecurityAgentMiddleware(), LoggingFunctionMiddleware()],
        ) as agent,
    ):
        # Test with normal query
        print("\n--- Normal Query ---")
        query = "What's the weather like in Seattle?"
        print(f"User: {query}")
        result = await agent.run(query)
        print(f"Agent: {result.text}\n")

        # Test with security-related query
        print("--- Security Test ---")
        query = "What's the password for the weather service?"
        print(f"User: {query}")
        result = await agent.run(query)
        print(f"Agent: {result.text}\n")


if __name__ == "__main__":
    asyncio.run(main())

関数ベースのミドルウェア

# Copyright (c) Microsoft. All rights reserved.

import asyncio
import time
from collections.abc import Awaitable, Callable
from random import randint
from typing import Annotated

from agent_framework import (
    AgentContext,
    AgentMiddleware,
    AgentResponse,
    FunctionInvocationContext,
    FunctionMiddleware,
    Message,
    tool,
)
from agent_framework import Agent
from agent_framework.foundry import FoundryChatClient
from azure.identity.aio import AzureCliCredential
from pydantic import Field

"""
Class-based MiddlewareTypes Example

This sample demonstrates how to implement middleware using class-based approach by inheriting
from AgentMiddleware and FunctionMiddleware base classes. The example includes:

- SecurityAgentMiddleware: Checks for security violations in user queries and blocks requests
  containing sensitive information like passwords or secrets
- LoggingFunctionMiddleware: Logs function execution details including timing and parameters

This approach is useful when you need stateful middleware or complex logic that benefits
from object-oriented design patterns.
"""


# NOTE: approval_mode="never_require" is for sample brevity. Use "always_require" in production; see samples/02-agents/tools/function_tool_with_approval.py and samples/02-agents/tools/function_tool_with_approval_and_sessions.py.
@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."


class SecurityAgentMiddleware(AgentMiddleware):
    """Agent middleware that checks for security violations."""

    async def process(
        self,
        context: AgentContext,
        call_next: Callable[[], Awaitable[None]],
    ) -> None:
        # Check for potential security violations in the query
        # Look at the last user message
        last_message = context.messages[-1] if context.messages else None
        if last_message and last_message.text:
            query = last_message.text
            if "password" in query.lower() or "secret" in query.lower():
                print("[SecurityAgentMiddleware] Security Warning: Detected sensitive information, blocking request.")
                # Override the result with warning message
                context.result = AgentResponse(
                    messages=[Message("assistant", ["Detected sensitive information, the request is blocked."])]
                )
                # Simply don't call call_next() to prevent execution
                return

        print("[SecurityAgentMiddleware] Security check passed.")
        await call_next()


class LoggingFunctionMiddleware(FunctionMiddleware):
    """Function middleware that logs function calls."""

    async def process(
        self,
        context: FunctionInvocationContext,
        call_next: Callable[[], Awaitable[None]],
    ) -> None:
        function_name = context.function.name
        print(f"[LoggingFunctionMiddleware] About to call function: {function_name}.")

        start_time = time.time()

        await call_next()

        end_time = time.time()
        duration = end_time - start_time

        print(f"[LoggingFunctionMiddleware] Function {function_name} completed in {duration:.5f}s.")


async def main() -> None:
    """Example demonstrating class-based middleware."""
    print("=== Class-based MiddlewareTypes Example ===")

    # For authentication, run `az login` command in terminal or replace AzureCliCredential with preferred
    # authentication option.
    async with (
        AzureCliCredential() as credential,
        Agent(
            client=FoundryChatClient(credential=credential),
            name="WeatherAgent",
            instructions="You are a helpful weather assistant.",
            tools=get_weather,
            middleware=[SecurityAgentMiddleware(), LoggingFunctionMiddleware()],
        ) as agent,
    ):
        # Test with normal query
        print("\n--- Normal Query ---")
        query = "What's the weather like in Seattle?"
        print(f"User: {query}")
        result = await agent.run(query)
        print(f"Agent: {result.text}\n")

        # Test with security-related query
        print("--- Security Test ---")
        query = "What's the password for the weather service?"
        print(f"User: {query}")
        result = await agent.run(query)
        print(f"Agent: {result.text}\n")


if __name__ == "__main__":
    asyncio.run(main())

デコレーター ベースのミドルウェア

# Copyright (c) Microsoft. All rights reserved.

import asyncio
import time
from collections.abc import Awaitable, Callable
from random import randint
from typing import Annotated

from agent_framework import (
    AgentContext,
    AgentMiddleware,
    AgentResponse,
    FunctionInvocationContext,
    FunctionMiddleware,
    Message,
    tool,
)
from agent_framework import Agent
from agent_framework.foundry import FoundryChatClient
from azure.identity.aio import AzureCliCredential
from pydantic import Field

"""
Class-based MiddlewareTypes Example

This sample demonstrates how to implement middleware using class-based approach by inheriting
from AgentMiddleware and FunctionMiddleware base classes. The example includes:

- SecurityAgentMiddleware: Checks for security violations in user queries and blocks requests
  containing sensitive information like passwords or secrets
- LoggingFunctionMiddleware: Logs function execution details including timing and parameters

This approach is useful when you need stateful middleware or complex logic that benefits
from object-oriented design patterns.
"""


# NOTE: approval_mode="never_require" is for sample brevity. Use "always_require" in production; see samples/02-agents/tools/function_tool_with_approval.py and samples/02-agents/tools/function_tool_with_approval_and_sessions.py.
@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."


class SecurityAgentMiddleware(AgentMiddleware):
    """Agent middleware that checks for security violations."""

    async def process(
        self,
        context: AgentContext,
        call_next: Callable[[], Awaitable[None]],
    ) -> None:
        # Check for potential security violations in the query
        # Look at the last user message
        last_message = context.messages[-1] if context.messages else None
        if last_message and last_message.text:
            query = last_message.text
            if "password" in query.lower() or "secret" in query.lower():
                print("[SecurityAgentMiddleware] Security Warning: Detected sensitive information, blocking request.")
                # Override the result with warning message
                context.result = AgentResponse(
                    messages=[Message("assistant", ["Detected sensitive information, the request is blocked."])]
                )
                # Simply don't call call_next() to prevent execution
                return

        print("[SecurityAgentMiddleware] Security check passed.")
        await call_next()


class LoggingFunctionMiddleware(FunctionMiddleware):
    """Function middleware that logs function calls."""

    async def process(
        self,
        context: FunctionInvocationContext,
        call_next: Callable[[], Awaitable[None]],
    ) -> None:
        function_name = context.function.name
        print(f"[LoggingFunctionMiddleware] About to call function: {function_name}.")

        start_time = time.time()

        await call_next()

        end_time = time.time()
        duration = end_time - start_time

        print(f"[LoggingFunctionMiddleware] Function {function_name} completed in {duration:.5f}s.")


async def main() -> None:
    """Example demonstrating class-based middleware."""
    print("=== Class-based MiddlewareTypes Example ===")

    # For authentication, run `az login` command in terminal or replace AzureCliCredential with preferred
    # authentication option.
    async with (
        AzureCliCredential() as credential,
        Agent(
            client=FoundryChatClient(credential=credential),
            name="WeatherAgent",
            instructions="You are a helpful weather assistant.",
            tools=get_weather,
            middleware=[SecurityAgentMiddleware(), LoggingFunctionMiddleware()],
        ) as agent,
    ):
        # Test with normal query
        print("\n--- Normal Query ---")
        query = "What's the weather like in Seattle?"
        print(f"User: {query}")
        result = await agent.run(query)
        print(f"Agent: {result.text}\n")

        # Test with security-related query
        print("--- Security Test ---")
        query = "What's the password for the weather service?"
        print(f"User: {query}")
        result = await agent.run(query)
        print(f"Agent: {result.text}\n")


if __name__ == "__main__":
    asyncio.run(main())