Поделиться через


Добавление промежуточного ПО для агентов

Узнайте, как добавить промежуточное ПО в своих агентов несколькими простыми шагами. Промежуточное ПО позволяет перехватывать и изменять взаимодействия агентов для ведения журнала, безопасности и решения других общих задач.

Предпосылки

Предварительные требования и установка пакетов NuGet см. в разделе "Создание и запуск простого агента " в этом руководстве.

Шаг 1. Создание простого агента

Сначала создайте базового агента с помощью инструмента функции.

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))]);

Предупреждение

DefaultAzureCredential удобно для разработки, но требует тщательного рассмотрения в рабочей среде. В рабочей среде рекомендуется использовать определенные учетные данные (например, ManagedIdentityCredential), чтобы избежать проблем с задержкой, непреднамеренной проверки данных аутентификации и потенциальных рисков безопасности из-за резервных механизмов.

Создайте Middleware для запуска вашего агента

Затем создайте функцию, которая будет вызываться для каждого запуска агента. Он позволяет проверять входные и выходные данные агента.

Если не планируется использовать ПО промежуточного слоя для остановки выполнения, функция должна вызывать RunAsync на указанном innerAgent.

Этот пример программного обеспечения промежуточного уровня просто анализирует входные и выходные данные выполнения агента и выводит количество сообщений, переданных в агент и из него.

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;
}

Шаг 3. Добавить Middleware для запуска агента в ваш агент

Чтобы добавить эту функцию промежуточного программного обеспечения в baseAgent, созданный на шаге 1, используйте шаблон построителя. При этом создается новый агент с примененным промежуточным программным обеспечением. Исходный файл baseAgent не изменяется.

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

Теперь, при выполнении запроса агентом, промежуточное ПО должно вызываться, отображая количество входных сообщений и количество ответных сообщений.

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

Шаг 4. Создание промежуточного программного слоя для вызова функции

Замечание

Промежуточное программное обеспечение для вызова функций в настоящее время поддерживается только с AIAgent, с использованием FunctionInvokingChatClient, например, ChatClientAgent.

Вы также можете создать посредническое ПО, которое вызывается при каждом вызове функции. Ниже приведен пример ПО промежуточного слоя для вызова функций, которое может проверять и/или изменять вызываемую функцию и результат вызова функции.

Если намерение заключается в том, чтобы не использовать промежуточное ПО для выполнения функции, промежуточное ПО должно вызвать предоставленный nextFunc.

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;
}

Шаг 5. Добавление программного обеспечения для вызова функции агенту

Так же, как и при добавлении связующего ПО, запущенного агентом, можно добавить связывающее ПО для вызова функций следующим образом:

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

Теперь при выполнении агента с запросом, вызывающим функцию, ПО промежуточного слоя должно вызываться, выводя имя функции и результат вызова.

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

Шаг 6. Создание ПО промежуточного слоя клиента чата

Для агентов, созданных с использованием IChatClient, может потребоваться перехват вызовов, поступающих от агента к агенту IChatClient. В этом случае можно использовать промежуточное ПО для IChatClient.

Ниже приведен пример промежуточного слоя клиента чата, который может проверять и/или изменять входные и выходные данные запроса к службе вывода данных, предоставляемой клиентом чата.

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;
}

Замечание

Для получения дополнительной информации о промежуточном ПО, смотрите IChatClient.

Шаг 7: Добавьте промежуточное ПО клиента чата в IChatClient

Чтобы добавить промежуточное ПО в ваш IChatClient, можно использовать шаблон построителя. После добавления промежуточного ПО IChatClient, вы можете использовать его с вашим агентом как обычно.

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 Промежуточное программное обеспечение также можно зарегистрировать, используя фабричный метод при создании агента с помощью одного из вспомогательных методов с клиентами SDK.

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());

Шаг 1. Создание простого агента

Сначала создайте базовый агент:

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())

Шаг 2. Создание промежуточного ПО

Создайте простое промежуточное ПО для ведения журнала, чтобы увидеть, когда выполняется ваш агент.

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!")

Шаг 3. Добавьте промежуточное ПО в вашего агента

Добавьте промежуточное ПО при создании вашего агента:

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)

Шаг 4: Создать промежуточное ПО для функций

Если агент использует функции, можно перехватывать вызовы функций и задавать значения среды выполнения только для инструментов перед выполнением средства:

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?")

Шаг 5. Использование ПО промежуточного слоя Run-Level

Вы также можете добавить ПО промежуточного слоя для определенных запусков:

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

Что дальше?

Для более сложных сценариев см. в Руководство пользователя по агентскому промежуточному программному обеспечению, в котором рассматривается:

  • Различные типы ПО промежуточного слоя (агент, функция, чат).
  • Программное обеспечение промежуточного слоя, основанное на классах, для сложных сценариев.
  • Завершение и переопределение посредника.
  • Продвинутые шаблоны промежуточного программного обеспечения и лучшие практики.

Полные примеры

Промежуточное ПО на основе классов

# 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())

Функциональная промежуточная платформа

# 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())

Промежуточное ПО на основе декораторов

# 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())

Дальнейшие шаги