Microsoft Foundry Início Rápido

1. Instale estes pacotes:

   pip install openai azure-identity azure-ai-projects==1.0.0
   

2. O Microsoft Foundry Models permite aos clientes consumir os modelos mais potentes dos fornecedores de modelos de referência usando um único endpoint e credenciais. Isso significa que você pode alternar entre modelos e consumi-los do seu aplicativo sem alterar uma única linha de código.

   Copie o endpoint do projeto Foundry na secção de Visão Geral do seu projeto. Você vai usá-lo em um momento.

   

   Tip

   Se não vires o endpoint do projeto Foundry, estás a usar um projeto baseado em hub. (Ver Tipos de projetos). Mude para um projeto do Foundry ou use as etapas anteriores para criar um.

   1. Selecione Página Inicial na navegação no canto superior direito.
   2. Selecione Chaves e copie o Endpoint. Você vai usá-lo em um momento.

3. Certifique-se de entrar usando o comando CLI az login (ou az login --use-device-code) para autenticar antes de executar seus scripts Python.

Acompanhe abaixo ou obtenha o código:

1. Instalar pacotes:

   Para trabalhar com o Foundry Tools no seu projeto .NET, terá de instalar vários pacotes NuGet. Adicione pacotes NuGet usando a CLI .NET no terminal integrado:

   # Add Azure AI SDK packages
   dotnet add package Azure.Identity
   dotnet add package Azure.AI.Projects 
   dotnet add package Azure.AI.Agents.Persistent
   dotnet add package Azure.AI.Inference
   

2. O Microsoft Foundry Models permite aos clientes consumir os modelos mais potentes dos fornecedores de modelos de referência usando um único endpoint e credenciais. Isso significa que você pode alternar entre modelos e consumi-los do seu aplicativo sem alterar uma única linha de código.

   Copie o endpoint do projeto Foundry na secção de Visão Geral do seu projeto. Você vai usá-lo em um momento.

   

   Tip

   Se não vires o endpoint do projeto Foundry, estás a usar um projeto baseado em hub. (Ver Tipos de projetos). Mude para um projeto do Foundry ou use as etapas anteriores para criar um.

   1. Selecione Página Inicial na navegação no canto superior direito.
   2. Selecione Chaves e copie o Endpoint. Você vai usá-lo em um momento.

3. Defina essas variáveis de ambiente para usar em seus scripts. O AZURE_AI_ENDPOINT é o ponto de extremidade do projeto que você copiou anteriormente. Remova tudo depois .com/ nesse ponto final para formar AZURE_AI_INFERENCE.

   AZURE_AI_ENDPOINT=https://your.services.ai.azure.com/api/projects/project
   AZURE_AI_INFERENCE=https://your.services.ai.azure.com/
   AZURE_AI_MODEL=your_model_name
   

   Tip

   As amostras de agente exigem que a AZURE_AI_MODEL variável de ambiente seja definida para um modelo compatível com OpenAI, por exemplo gpt-4.1, pois nem todos os modelos são suportados para casos de uso do agente, incluindo ferramentas.

4. Certifique-se de entrar usando o comando CLI az login (ou az login --use-device-code) para autenticar antes de executar seus scripts C#.

Acompanhe abaixo ou obtenha o código:

1. Certifique-se de entrar usando o comando CLI az login (ou az login --use-device-code) para autenticar antes de executar seus scripts TypeScript.

2. Baixar package.json.

3. Instale pacotes com npm install

4. O Microsoft Foundry Models permite aos clientes consumir os modelos mais potentes dos fornecedores de modelos de referência usando um único endpoint e credenciais. Isso significa que você pode alternar entre modelos e consumi-los do seu aplicativo sem alterar uma única linha de código.

   Copie o endpoint do projeto Foundry na secção de Visão Geral do seu projeto. Você vai usá-lo em um momento.

   

   Tip

   Se não vires o endpoint do projeto Foundry, estás a usar um projeto baseado em hub. (Ver Tipos de projetos). Mude para um projeto do Foundry ou use as etapas anteriores para criar um.

   1. Selecione Página Inicial na navegação no canto superior direito.
   2. Selecione Chaves e copie o Endpoint. Você vai usá-lo em um momento.

5. Defina estas variáveis de ambiente para usar em seus scripts:

   MODEL_DEPLOYMENT_NAME=gpt-4o
   PROJECT_ENDPOINT=https://<your-foundry-resource-name>.services.ai.azure.com/api/projects/<your-foundry-project-name>
   

6. Comece o seu código com estas importações:

   import fs from 'fs';
   import path from 'path';
   import { fileURLToPath } from 'url';
   import { DefaultAzureCredential } from '@azure/identity';
   import { ToolUtility, DoneEvent, ErrorEvent, ThreadMessage } from '@azure/ai-agents';
   import { AIProjectClient } from '@azure/ai-projects';
   import { config } from 'dotenv';
   config();
   

Acompanhe abaixo ou obtenha o código:

1. O Microsoft Foundry Models permite aos clientes consumir os modelos mais potentes dos fornecedores de modelos de referência usando um único endpoint e credenciais. Isso significa que você pode alternar entre modelos e consumi-los do seu aplicativo sem alterar uma única linha de código.

   Copie o endpoint do projeto Foundry na secção de Visão Geral do seu projeto. Você vai usá-lo em um momento.

   

   Tip

   Se não vires o endpoint do projeto Foundry, estás a usar um projeto baseado em hub. (Ver Tipos de projetos). Mude para um projeto do Foundry ou use as etapas anteriores para criar um.

   1. Selecione Página Inicial na navegação no canto superior direito.
   2. Selecione Chaves e copie o Endpoint. Você vai usá-lo em um momento.

2. Defina estas variáveis de ambiente para usar em seus scripts:

   MODEL_DEPLOYMENT_NAME=gpt-4o
   PROJECT_ENDPOINT=https://<your-foundry-resource-name>.services.ai.azure.com/api/projects/<your-foundry-project-name>
   

3. Certifique-se de fazer login usando o comando CLI az login (ou az login --use-device-code) para autenticar antes de executar seus scripts Java.

4. Faça o download POM.XML para o seu IDE Java.

Acompanhe abaixo ou obtenha o código:

1. Certifique-se de entrar usando o comando CLI az login (ou az login --use-device-code) para autenticar antes de executar o próximo comando.

2. Obtenha um token de acesso temporário. Ele expirará em 60-90 minutos, você precisará atualizar depois disso.

   az account get-access-token --scope https://ai.azure.com/.default
   

3. Salve os resultados como a variável AZURE_AI_AUTH_TOKENde ambiente .

Acompanhe abaixo ou obtenha o código:

Não é necessária qualquer instalação para utilizar o portal da Foundry.

1. Instale estes pacotes, incluindo a versão de pré-visualização de azure-ai-projects. Esta versão utiliza a API (nova) dos projetos Foundry (pré-visualização).

   pip install azure-ai-projects --pre
   pip install openai azure-identity python-dotenv
   

2. Certifique-se de entrar usando o comando CLI az login (ou az login --use-device-code) para autenticar antes de executar seus scripts Python.

Acompanhe abaixo ou obtenha o código:

1. Instalar pacotes:

   Adicionar pacotes NuGet usando a CLI .NET no terminal integrado: Estes pacotes usam a API (nova) dos projetos Foundry (pré-visualização).

   dotnet add package Azure.AI.Projects --prerelease
   dotnet add package Azure.AI.Projects.OpenAI --prerelease
   dotnet add package Azure.Identity
   

2. Certifique-se de entrar usando o comando CLI az login (ou az login --use-device-code) para autenticar antes de executar seus scripts C#.

Acompanhe abaixo ou obtenha o código:

1. Instale estes pacotes, incluindo a versão de pré-visualização de @azure/ai-projects. Esta versão utiliza a API (nova) dos projetos Foundry (pré-visualização).:

   npm install @azure/ai-projects@beta @azure/identity dotenv
   

2. Certifique-se de entrar usando o comando CLI az login (ou az login --use-device-code) para autenticar antes de executar seus scripts TypeScript.

Acompanhe abaixo ou obtenha o código:

1. Certifique-se de fazer login usando o comando CLI az login (ou az login --use-device-code) para autenticar antes de executar seus scripts Java.

Acompanhe abaixo ou obtenha o código:

1. Certifique-se de entrar usando o comando CLI az login (ou az login --use-device-code) para autenticar antes de executar o próximo comando.

2. Obtenha um token de acesso temporário. Ele expirará em 60-90 minutos, você precisará atualizar depois disso.

   az account get-access-token --scope https://ai.azure.com/.default
   

3. Salve os resultados como a variável AZURE_AI_AUTH_TOKENde ambiente .

Acompanhe abaixo ou obtenha o código:

Não é necessária qualquer instalação para utilizar o portal da Foundry.

Substitua pelo endpoint o seu ponto de extremidade neste código.

from azure.ai.projects import AIProjectClient
from azure.identity import DefaultAzureCredential

project = AIProjectClient(
    endpoint="https://your-foundry-resource-name.ai.azure.com/api/projects/project-name",
    credential=DefaultAzureCredential(),
)

models = project.get_openai_client(api_version="2024-10-21")
response = models.chat.completions.create(
    model="gpt-4o",
    messages=[
        {"role": "system", "content": "You are a helpful writing assistant"},
        {"role": "user", "content": "Write me a poem about flowers"},
    ],
)

print(response.choices[0].message.content)

using System.ClientModel.Primitives;
using Azure.Identity;
using OpenAI;
using OpenAI.Chat;

#pragma warning disable OPENAI001

string projectEndpoint = System.Environment.GetEnvironmentVariable("AZURE_AI_INFERENCE")!;
string modelDeploymentName = System.Environment.GetEnvironmentVariable("AZURE_AI_MODEL")!;

BearerTokenPolicy tokenPolicy = new(
    new DefaultAzureCredential(),
    "https://ai.azure.com/.default");
OpenAIClient openAIClient = new(
    authenticationPolicy: tokenPolicy,
    options: new OpenAIClientOptions()
    {
        Endpoint = new($"{projectEndpoint}/openai/v1"),
    });
ChatClient chatClient = openAIClient.GetChatClient(modelDeploymentName);

ChatCompletion completion = await chatClient.CompleteChatAsync(
    [
        new SystemChatMessage("You are a helpful assistant."),
                    new UserChatMessage("How many feet are in a mile?")
    ]);

Console.WriteLine(completion.Content[0].Text);

// Get the Azure AI endpoint and deployment name from environment variables
const endpoint = process.env.PROJECT_ENDPOINT as string;
const deployment = process.env.MODEL_DEPLOYMENT_NAME || 'gpt-4o';

// Create an Azure OpenAI Client
const project = new AIProjectClient(endpoint, new DefaultAzureCredential());
const client = await project.getAzureOpenAIClient({
    // The API version should match the version of the Azure OpenAI resource
    apiVersion: "2024-12-01-preview"
});

// Create a chat completion
const chatCompletion = await client.chat.completions.create({
    model: deployment,
    messages: [
        { role: "system", content: "You are a helpful writing assistant" },
        { role: "user", content: "Write me a poem about flowers" },
    ],
});
console.log(`\n==================== 🌷 COMPLETIONS POEM ====================\n`);
console.log(chatCompletion.choices[0].message.content);

package com.azure.ai.foundry.samples;

import com.azure.ai.inference.ChatCompletionsClient;
import com.azure.ai.inference.ChatCompletionsClientBuilder;
import com.azure.ai.inference.models.ChatCompletions;
import com.azure.core.credential.AzureKeyCredential;
import com.azure.core.credential.TokenCredential;
import com.azure.core.exception.HttpResponseException;
import com.azure.core.util.logging.ClientLogger;
import com.azure.identity.DefaultAzureCredentialBuilder;

/**
 * Sample demonstrating non-streaming chat completion functionality
 * using the Azure AI Inference SDK, wired to your AOAI project endpoint.
 *
 * Environment variables:
 * - PROJECT_ENDPOINT:          Required. Your Azure AI project endpoint.
 * - AZURE_AI_API_KEY:          Optional. Your API key (falls back to DefaultAzureCredential).
 * - AZURE_MODEL_DEPLOYMENT_NAME: Optional. Model deployment name (default: "phi-4").
 * - AZURE_MODEL_API_PATH:      Optional. API path segment (default: "deployments").
 * - CHAT_PROMPT:               Optional. The prompt to send (uses a default if not provided).
 * 
 * SDK Features Demonstrated:
 * - Using the Azure AI Inference SDK (com.azure:azure-ai-inference:1.0.0-beta.5)
 * - Creating a ChatCompletionsClient with Azure or API key authentication
 * - Configuring endpoint paths for different model deployments
 * - Using the simplified complete() method for quick completions
 * - Accessing response content through strongly-typed objects
 * - Implementing proper error handling for service requests
 * - Choosing between DefaultAzureCredential and AzureKeyCredential
 * 
 */

public class ChatCompletionSample {
    private static final ClientLogger logger = new ClientLogger(ChatCompletionSample.class);

    public static void main(String[] args) {
        // 1) Read and validate the project endpoint
        String projectEndpoint = System.getenv("PROJECT_ENDPOINT");
        if (projectEndpoint == null || projectEndpoint.isBlank()) {
            logger.error("PROJECT_ENDPOINT is required but not set");
            return;
        }

        // 2) Optional auth + model settings
        String apiKey          = System.getenv("AZURE_AI_API_KEY");
        String deploymentName  = System.getenv("AZURE_MODEL_DEPLOYMENT_NAME");
        String apiPath         = System.getenv("AZURE_MODEL_API_PATH");
        String prompt          = System.getenv("CHAT_PROMPT");

        if (deploymentName == null || deploymentName.isBlank()) {
            deploymentName = "phi-4";
            logger.info("No AZURE_MODEL_DEPLOYMENT_NAME provided, using default: {}", deploymentName);
        }
        if (apiPath == null || apiPath.isBlank()) {
            apiPath = "deployments";
            logger.info("No AZURE_MODEL_API_PATH provided, using default: {}", apiPath);
        }
        if (prompt == null || prompt.isBlank()) {
            prompt = "What best practices should I follow when asking an AI model to review Java code?";
            logger.info("No CHAT_PROMPT provided, using default prompt: {}", prompt);
        }

        try {
            // 3) Build the full inference endpoint URL
            String fullEndpoint = projectEndpoint.endsWith("/") 
                ? projectEndpoint 
                : projectEndpoint + "/";
            fullEndpoint += apiPath + "/" + deploymentName;
            logger.info("Using inference endpoint: {}", fullEndpoint);

            // 4) Create the client with key or token credential :contentReference[oaicite:0]{index=0}
            ChatCompletionsClient client;
            if (apiKey != null && !apiKey.isBlank()) {
                logger.info("Authenticating using API key");
                client = new ChatCompletionsClientBuilder()
                    .credential(new AzureKeyCredential(apiKey))
                    .endpoint(fullEndpoint)
                    .buildClient();
            } else {
                logger.info("Authenticating using DefaultAzureCredential");
                TokenCredential credential = new DefaultAzureCredentialBuilder().build();
                client = new ChatCompletionsClientBuilder()
                    .credential(credential)
                    .endpoint(fullEndpoint)
                    .buildClient();
            }

            // 5) Send a simple chat completion request
            logger.info("Sending chat completion request with prompt: {}", prompt);
            ChatCompletions completions = client.complete(prompt);

            // 6) Process the response
            String content = completions.getChoice().getMessage().getContent();
            logger.info("Received response from model");
            System.out.println("\nResponse from AI assistant:\n" + content);
        } catch (HttpResponseException e) {
            // Handle API errors
            int status = e.getResponse().getStatusCode();
            logger.error("Service error {}: {}", status, e.getMessage());
            if (status == 401 || status == 403) {
                logger.error("Authentication failed. Check API key or Azure credentials.");
            } else if (status == 404) {
                logger.error("Deployment not found. Verify deployment name and endpoint.");
            } else if (status == 429) {
                logger.error("Rate limit exceeded. Please retry later.");
            }
        } catch (Exception e) {
            // Handle all other exceptions
            logger.error("Error in chat completion: {}", e.getMessage(), e);
        }
    }
}

Substitua YOUR-FOUNDRY-RESOURCE-NAME pelos seus valores:

curl --request POST --url 'https://YOUR-FOUNDRY-RESOURCE-NAME.services.ai.azure.com/openai/deployments/gpt-4o/chat/completions?api-version=2024-10-21' \
-h 'authorization: Bearer $AZURE_AI_AUTH_TOKEN' \
-h 'content-type: application/json' \
-d '{
        "messages": [
            {"role": "system", 
            "content": "You are a helpful writing assistant"},
            {"role": "user", 
            "content": "Write me a poem about flowers"}
        ],
        "model": "gpt-4o"
    }'

1. Na área de teste do chat, preencha o campo de entrada e selecione Enviar.
2. O modelo retorna uma resposta no painel Resposta .

import os
from dotenv import load_dotenv
from azure.identity import DefaultAzureCredential
from azure.ai.projects import AIProjectClient

load_dotenv()

print(f"Using PROJECT_ENDPOINT: {os.environ['PROJECT_ENDPOINT']}")
print(f"Using MODEL_DEPLOYMENT_NAME: {os.environ['MODEL_DEPLOYMENT_NAME']}")

project_client = AIProjectClient(
    endpoint=os.environ["PROJECT_ENDPOINT"],
    credential=DefaultAzureCredential(),
)

openai_client = project_client.get_openai_client()

response = openai_client.responses.create(
    model=os.environ["MODEL_DEPLOYMENT_NAME"],
    input="What is the size of France in square miles?",
)
print(f"Response output: {response.output_text}")

using Azure.AI.Projects;
using Azure.AI.Projects.OpenAI;
using Azure.Identity;
using OpenAI;
using OpenAI.Responses;

#pragma warning disable OPENAI001

string projectEndpoint  = Environment.GetEnvironmentVariable("PROJECT_ENDPOINT")
    ?? throw new InvalidOperationException("Missing environment variable 'PROJECT_ENDPOINT'");
string modelDeploymentName  = Environment.GetEnvironmentVariable("MODEL_DEPLOYMENT_NAME")
    ?? throw new InvalidOperationException("Missing environment variable 'MODEL_DEPLOYMENT_NAME'");

AIProjectClient projectClient = new(new Uri(projectEndpoint ), new AzureCliCredential());

ProjectResponsesClient responseClient = projectClient.OpenAI.GetProjectResponsesClientForModel(modelDeploymentName);
ResponseResult response = await responseClient.CreateResponseAsync("What is the size of France in square miles?");

Console.WriteLine(response.GetOutputText());

import { DefaultAzureCredential } from "@azure/identity";
import { AIProjectClient } from "@azure/ai-projects";
import "dotenv/config";

const projectEndpoint = process.env["PROJECT_ENDPOINT"] || "<project endpoint>";
const deploymentName = process.env["MODEL_DEPLOYMENT_NAME"] || "<model deployment name>";

async function main(): Promise<void> {
    const project = new AIProjectClient(projectEndpoint, new DefaultAzureCredential());
    const openAIClient = await project.getOpenAIClient();
    const response = await openAIClient.responses.create({
        model: deploymentName,
        input: "What is the size of France in square miles?",
    });
    console.log(`Response output: ${response.output_text}`);
}

main().catch(console.error);

package com.azure.ai.agents;

import com.azure.core.util.Configuration;
import com.azure.identity.DefaultAzureCredentialBuilder;
import com.openai.models.responses.Response;
import com.openai.models.responses.ResponseCreateParams;

public class CreateResponse {
    public static void main(String[] args) {
        String endpoint = Configuration.getGlobalConfiguration().get("PROJECT_ENDPOINT");
        String model = Configuration.getGlobalConfiguration().get("MODEL_DEPLOYMENT_NAME");
        // Code sample for creating a response
        ResponsesClient responsesClient = new AgentsClientBuilder()
                .credential(new DefaultAzureCredentialBuilder().build())
                .endpoint(endpoint)
                .serviceVersion(AgentsServiceVersion.V2025_11_15_PREVIEW)
                .buildResponsesClient();

        ResponseCreateParams responseRequest = new ResponseCreateParams.Builder()
                .input("Hello, how can you help me?")
                .model(model)
                .build();

        Response response = responsesClient.getResponseService().create(responseRequest);

        System.out.println("Response ID: " + response.id());
        System.out.println("Response Model: " + response.model());
        System.out.println("Response Created At: " + response.createdAt());
        System.out.println("Response Output: " + response.output());
    }
}

Substitua YOUR-FOUNDRY-RESOURCE-NAME pelos seus valores:

curl -X POST https://YOUR-FOUNDRY-RESOURCE-NAME.services.ai.azure.com/api/projects/YOUR-PROJECT-NAME/openai/responses?api-version=2025-11-15-preview \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $AZURE_AI_AUTH_TOKEN" \
-d '{
        "model": "gpt-4.1-mini",
        "input": "What is the size of France in square miles?"
}'

1. Depois que o modelo for implantado, você será automaticamente movido de Início para a seção Compilação . O seu novo modelo está selecionado e pronto para experimentar.

2. Comece a conversar com seu modelo, por exemplo: "Escreva-me um poema sobre flores".

import os
from dotenv import load_dotenv
from azure.identity import DefaultAzureCredential
from azure.ai.projects import AIProjectClient
from azure.ai.projects.models import PromptAgentDefinition

load_dotenv()

project_client = AIProjectClient(
    endpoint=os.environ["PROJECT_ENDPOINT"],
    credential=DefaultAzureCredential(),
)

agent = project_client.agents.create_version(
    agent_name=os.environ["AGENT_NAME"],
    definition=PromptAgentDefinition(
        model=os.environ["MODEL_DEPLOYMENT_NAME"],
        instructions="You are a helpful assistant that answers general questions",
    ),
)
print(f"Agent created (id: {agent.id}, name: {agent.name}, version: {agent.version})")

using Azure.AI.Projects;
using Azure.AI.Projects.OpenAI;
using Azure.Identity;

string projectEndpoint = Environment.GetEnvironmentVariable("PROJECT_ENDPOINT")
    ?? throw new InvalidOperationException("Missing environment variable 'PROJECT_ENDPOINT'");
string modelDeploymentName = Environment.GetEnvironmentVariable("MODEL_DEPLOYMENT_NAME")
    ?? throw new InvalidOperationException("Missing environment variable 'MODEL_DEPLOYMENT_NAME'");
string agentName = Environment.GetEnvironmentVariable("AGENT_NAME")
    ?? throw new InvalidOperationException("Missing environment variable 'AGENT_NAME'");

AIProjectClient projectClient = new(new Uri(projectEndpoint), new AzureCliCredential());

AgentDefinition agentDefinition = new PromptAgentDefinition(modelDeploymentName)
{
    Instructions = "You are a helpful assistant that answers general questions",
};

AgentVersion newAgentVersion = projectClient.Agents.CreateAgentVersion(
    agentName,
    options: new(agentDefinition));

List<AgentVersion> agentVersions = projectClient.Agents.GetAgentVersions(agentName);
foreach (AgentVersion agentVersion in agentVersions)
{
    Console.WriteLine($"Agent: {agentVersion.Id}, Name: {agentVersion.Name}, Version: {agentVersion.Version}");
}

import { DefaultAzureCredential } from "@azure/identity";
import { AIProjectClient } from "@azure/ai-projects";
import "dotenv/config";

const projectEndpoint = process.env["PROJECT_ENDPOINT"] || "<project endpoint>";
const deploymentName = process.env["MODEL_DEPLOYMENT_NAME"] || "<model deployment name>";

async function main(): Promise<void> {
    const project = new AIProjectClient(projectEndpoint, new DefaultAzureCredential());
    const agent = await project.agents.createVersion("my-agent-basic", {
        kind: "prompt",
        model: deploymentName,
        instructions: "You are a helpful assistant that answers general questions",
  });
  console.log(`Agent created (id: ${agent.id}, name: ${agent.name}, version: ${agent.version})`);
}

main().catch(console.error);

package com.azure.ai.agents;

import com.azure.ai.agents.models.AgentVersionDetails;
import com.azure.ai.agents.models.PromptAgentDefinition;
import com.azure.core.util.Configuration;
import com.azure.identity.DefaultAzureCredentialBuilder;

public class CreateAgent {
    public static void main(String[] args) {
        String endpoint = Configuration.getGlobalConfiguration().get("PROJECT_ENDPOINT");
        String model = Configuration.getGlobalConfiguration().get("MODEL_DEPLOYMENT_NAME");
        // Code sample for creating an agent
        AgentsClient agentsClient = new AgentsClientBuilder()
                .credential(new DefaultAzureCredentialBuilder().build())
                .endpoint(endpoint)
                .buildAgentsClient();

        PromptAgentDefinition request = new PromptAgentDefinition(model);
        AgentVersionDetails agent = agentsClient.createAgentVersion("MyAgent", request);

        System.out.println("Agent ID: " + agent.getId());
        System.out.println("Agent Name: " + agent.getName());
        System.out.println("Agent Version: " + agent.getVersion());
    }
}

Substitua YOUR-FOUNDRY-RESOURCE-NAME pelos seus valores:

curl -X POST https://YOUR-FOUNDRY-RESOURCE-NAME.services.ai.azure.com/api/projects/YOUR-PROJECT-NAME/agents?api-version=2025-11-15-preview \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $AZURE_AI_AUTH_TOKEN" \
  -d '{
        "name": "MyAgent",
        "definition": {
            "kind": "prompt",
            "model": "gpt-4.1-mini", 
            "instructions": "You are a helpful assistant that answers general questions"
        }
    }'

Agora crie um agente e interaja com ele.

1. Ainda na seção Compilar , selecione Agentes no painel esquerdo.
2. Selecione Criar agente e atribua-lhe um nome.

Substitua pelo endpoint o seu ponto de extremidade neste código.

from azure.ai.projects import AIProjectClient
from azure.identity import DefaultAzureCredential
from azure.ai.agents.models import ListSortOrder, FilePurpose

project = AIProjectClient(
    endpoint="https://your-foundry-resource-name.ai.azure.com/api/projects/project-name",
    credential=DefaultAzureCredential(),
)

agent = project.agents.create_agent(
    model="gpt-4o",
    name="my-agent",
    instructions="You are a helpful writing assistant")

thread = project.agents.threads.create()
message = project.agents.messages.create(
    thread_id=thread.id, 
    role="user", 
    content="Write me a poem about flowers")

run = project.agents.runs.create_and_process(thread_id=thread.id, agent_id=agent.id)
if run.status == "failed":
    # Check if you got "Rate limit is exceeded.", then you want to get more quota
    print(f"Run failed: {run.last_error}")

# Get messages from the thread
messages = project.agents.messages.list(thread_id=thread.id)

# Get the last message from the sender
messages = project.agents.messages.list(thread_id=thread.id, order=ListSortOrder.ASCENDING)
for message in messages:
    if message.run_id == run.id and message.text_messages:
        print(f"{message.role}: {message.text_messages[-1].text.value}")

# Delete the agent once done
project.agents.delete_agent(agent.id)
print("Deleted agent")

using Azure;
using Azure.Identity;
using Azure.AI.Agents.Persistent;

// Creating the Client for agents
var projectEndpoint = System.Environment.GetEnvironmentVariable("AZURE_AI_ENDPOINT");
var modelDeploymentName = System.Environment.GetEnvironmentVariable("AZURE_AI_MODEL");
PersistentAgentsClient client = new(projectEndpoint, new DefaultAzureCredential());

// Create an Agent with toolResources and process Agent run
PersistentAgent agent = client.Administration.CreateAgent(
        model: modelDeploymentName,
        name: "SDK Test Agent - Tutor",
        instructions: "You are a personal electronics tutor. Write and run code to answer questions.",
        tools: new List<ToolDefinition> { new CodeInterpreterToolDefinition() });

// Create thread for communication
PersistentAgentThread thread = client.Threads.CreateThread();

// Create message to thread
PersistentThreadMessage messageResponse = client.Messages.CreateMessage(
    thread.Id,
    MessageRole.User,
    "I need to solve the equation `3x + 11 = 14`. Can you help me?");

// Run the Agent
ThreadRun run = client.Runs.CreateRun(thread, agent);

// Wait for the run to complete
do
{
    Thread.Sleep(TimeSpan.FromMilliseconds(500));
    run = client.Runs.GetRun(thread.Id, run.Id);
}
while (run.Status == RunStatus.Queued
    || run.Status == RunStatus.InProgress);
Pageable<PersistentThreadMessage> messages = client.Messages.GetMessages(
    threadId: thread.Id,
    order: ListSortOrder.Ascending
);
// Print the messages in the thread
WriteMessages(messages);

// Delete the thread and agent after use
client.Threads.DeleteThread(thread.Id);
client.Administration.DeleteAgent(agent.Id);

// Temporary function to use a list of messages in the thread and write them to the console.
static void WriteMessages(IEnumerable<PersistentThreadMessage> messages)
{
    foreach (PersistentThreadMessage threadMessage in messages)
    {
        Console.Write($"{threadMessage.CreatedAt:yyyy-MM-dd HH:mm:ss} - {threadMessage.Role,10}: ");
        foreach (MessageContent contentItem in threadMessage.ContentItems)
        {
            if (contentItem is MessageTextContent textItem)
            {
                Console.Write(textItem.Text);
            }
            else if (contentItem is MessageImageFileContent imageFileItem)
            {
                Console.Write($"<image from ID: {imageFileItem.FileId}");
            }
            Console.WriteLine();
        }
    }
}

const endpoint = process.env.PROJECT_ENDPOINT as string;
const deployment = process.env.MODEL_DEPLOYMENT_NAME || 'gpt-4o';
const client = new AIProjectClient(endpoint, new DefaultAzureCredential());

// Create an Agent
const agent = await client.agents.createAgent(deployment, {
    name: 'my-agent',
    instructions: 'You are a helpful agent'
});
console.log(`\n==================== 🕵️  POEM AGENT ====================`);

// Create a thread and message
const thread = await client.agents.threads.create();
const prompt = 'Write me a poem about flowers';
console.log(`\n---------------- 📝 User Prompt ---------------- \n${prompt}`);
await client.agents.messages.create(thread.id, 'user', prompt);

// Create run
let run = await client.agents.runs.create(thread.id, agent.id);

// Wait for run to complete
console.log(`\n---------------- 🚦 Run Status ----------------`);
while (['queued', 'in_progress', 'requires_action'].includes(run.status)) {
    // Avoid adding a lot of messages to the console
    await new Promise((resolve) => setTimeout(resolve, 1000));
    run = await client.agents.runs.get(thread.id, run.id);
    console.log(`Run status: ${run.status}`);
}

console.log('\n---------------- 📊 Token Usage ----------------');
console.table([run.usage]);

const messagesIterator = await client.agents.messages.list(thread.id);
const assistantMessage = await getAssistantMessage(messagesIterator);
console.log('\n---------------- 💬 Response ----------------');
printAssistantMessage(assistantMessage);

// Clean up
console.log(`\n---------------- 🧹 Clean Up Poem Agent ----------------`);
await client.agents.deleteAgent(agent.id);
console.log(`Deleted Agent, Agent ID: ${agent.id}`);

package com.azure.ai.foundry.samples;

import com.azure.ai.agents.persistent.PersistentAgentsClient;
import com.azure.ai.agents.persistent.PersistentAgentsClientBuilder;
import com.azure.ai.agents.persistent.PersistentAgentsAdministrationClient;
import com.azure.ai.agents.persistent.models.CreateAgentOptions;
import com.azure.ai.agents.persistent.models.CreateThreadAndRunOptions;
import com.azure.ai.agents.persistent.models.PersistentAgent;
import com.azure.ai.agents.persistent.models.ThreadRun;
import com.azure.core.credential.TokenCredential;
import com.azure.core.exception.HttpResponseException;
import com.azure.core.util.logging.ClientLogger;
import com.azure.identity.DefaultAzureCredentialBuilder;


/**
 * Sample demonstrating how to work with Azure AI Agents using the Azure AI Agents Persistent SDK.
 * 
 * This sample shows how to:
 * - Set up authentication with Azure credentials
 * - Create a persistent agent with custom instructions
 * - Start a thread and run with the agent
 * - Access various properties of the agent and thread run
 * - Work with the PersistentAgentsClient and PersistentAgentsAdministrationClient
 * 
 * Environment variables:
 * - AZURE_ENDPOINT: Optional fallback. The base endpoint for your Azure AI service if PROJECT_ENDPOINT is not provided.
 * - PROJECT_ENDPOINT: Required. The endpoint for your Azure AI Project.
 * - MODEL_DEPLOYMENT_NAME: Optional. The model deployment name (defaults to "gpt-4o").
 * - AGENT_NAME: Optional. The name to give to the created agent (defaults to "java-quickstart-agent").
 * - AGENT_INSTRUCTIONS: Optional. The instructions for the agent (defaults to a helpful assistant).
 * 
 * Note: This sample requires proper Azure authentication. It uses DefaultAzureCredential which supports
 * multiple authentication methods including environment variables, managed identities, and interactive login.
 * 
 * SDK Features Demonstrated:
 * - Using the Azure AI Agents Persistent SDK (com.azure:azure-ai-agents-persistent:1.0.0-beta.2)
 * - Creating an authenticated client with DefaultAzureCredential
 * - Using the PersistentAgentsClientBuilder pattern for client instantiation
 * - Working with the PersistentAgentsAdministrationClient for agent management
 * - Creating agents with specific configurations (name, model, instructions)
 * - Starting threads and runs for agent conversations
 * - Working with agent state and thread management
 * - Accessing agent and thread run properties
 * - Implementing proper error handling for Azure service interactions
 */
public class AgentSample {
    private static final ClientLogger logger = new ClientLogger(AgentSample.class);

    public static void main(String[] args) {
        // Load environment variables with better error handling, supporting both .env and system environment variables
        String endpoint = System.getenv("AZURE_ENDPOINT");
        String projectEndpoint = System.getenv("PROJECT_ENDPOINT");
        String modelName = System.getenv("MODEL_DEPLOYMENT_NAME");
        String agentName = System.getenv("AGENT_NAME");
        String instructions = System.getenv("AGENT_INSTRUCTIONS");

        

        // Check for required endpoint configuration
        if (projectEndpoint == null && endpoint == null) {
            String errorMessage = "Environment variables not configured. Required: either PROJECT_ENDPOINT or AZURE_ENDPOINT must be set.";
            logger.error("ERROR: {}", errorMessage);
            logger.error("Please set your environment variables or create a .env file. See README.md for details.");
            return;
        }
        
        // Use AZURE_ENDPOINT as fallback if PROJECT_ENDPOINT not set
        if (projectEndpoint == null) {
            projectEndpoint = endpoint;
            logger.info("Using AZURE_ENDPOINT as PROJECT_ENDPOINT: {}", projectEndpoint);
        }

        // Set defaults for optional parameters with informative logging
        if (modelName == null) {
            modelName = "gpt-4o";
            logger.info("No MODEL_DEPLOYMENT_NAME provided, using default: {}", modelName);
        }
        if (agentName == null) {
            agentName = "java-quickstart-agent";
            logger.info("No AGENT_NAME provided, using default: {}", agentName);
        }
        if (instructions == null) {
            instructions = "You are a helpful assistant that provides clear and concise information.";
            logger.info("No AGENT_INSTRUCTIONS provided, using default instructions");
        }

        // Create Azure credential with DefaultAzureCredentialBuilder
        // This supports multiple authentication methods including environment variables,
        // managed identities, and interactive browser login
        logger.info("Building DefaultAzureCredential");
        TokenCredential credential = new DefaultAzureCredentialBuilder().build();

        try {
            // Build the general agents client
            logger.info("Creating PersistentAgentsClient with endpoint: {}", projectEndpoint);
            PersistentAgentsClient agentsClient = new PersistentAgentsClientBuilder()
                .endpoint(projectEndpoint)
                .credential(credential)
                .buildClient();

            // Derive the administration client
            logger.info("Getting PersistentAgentsAdministrationClient");
            PersistentAgentsAdministrationClient adminClient =
                agentsClient.getPersistentAgentsAdministrationClient();

            // Create an agent
            logger.info("Creating agent with name: {}, model: {}", agentName, modelName);
            PersistentAgent agent = adminClient.createAgent(
                new CreateAgentOptions(modelName)
                    .setName(agentName)
                    .setInstructions(instructions)
            );
            logger.info("Agent created: ID={}, Name={}", agent.getId(), agent.getName());
            logger.info("Agent model: {}", agent.getModel());

            // Start a thread/run on the general client
            logger.info("Creating thread and run with agent ID: {}", agent.getId());
            ThreadRun runResult = agentsClient.createThreadAndRun(
                new CreateThreadAndRunOptions(agent.getId())
            );
            logger.info("ThreadRun created: ThreadId={}", runResult.getThreadId());

            // List available getters on ThreadRun for informational purposes
            logger.info("\nAvailable getters on ThreadRun:");
            for (var method : ThreadRun.class.getMethods()) {
                if (method.getName().startsWith("get")) {
                    logger.info(" - {}", method.getName());
                }
            }

            logger.info("\nDemo completed successfully!");
            
        } catch (HttpResponseException e) {
            // Handle service-specific errors with detailed information
            int statusCode = e.getResponse().getStatusCode();
            logger.error("Service error {}: {}", statusCode, e.getMessage());
            logger.error("Refer to the Azure AI Agents documentation for troubleshooting information.");
        } catch (Exception e) {
            // Handle general exceptions
            logger.error("Error in agent sample: {}", e.getMessage(), e);
        }
    }
}

Substitua YOUR-FOUNDRY-RESOURCE-NAME e YOUR-PROJECT-NAME pelos seus valores:

# Create agent
curl --request POST --url "https://YOUR-FOUNDRY-RESOURCE-NAME.services.ai.azure.com/api/projects/YOUR-PROJECT-NAME/assistants?api-version=v1" \
    -h "authorization: Bearer $AZURE_AI_AUTH_TOKEN" \
    -h "content-type: application/json" \
    -d '{
            "model": "gpt-4o",
            "name": "my-agent",
            "instructions": "You are a helpful writing assistant"
        }'
#Lets say agent ID created is asst_123456789. Use this to run the agent

# Create thread
curl --request POST --url 'https://YOUR-FOUNDRY-RESOURCE-NAME.services.ai.azure.com/api/projects/YOUR-PROJECT-NAME/threads?api-version=v1' \
    -h 'authorization: Bearer $AZURE_AI_AUTH_TOKEN' \
    -h 'content-type: application/json'

#Lets say thread ID created is thread_123456789. Use this in the next step

# Create message using thread ID
curl --request POST --url 'https://YOUR-FOUNDRY-RESOURCE-NAME.services.ai.azure.com/api/projects/YOUR-PROJECT-NAME/threads/thread_123456789/messages?api-version=v1' \
    -h 'authorization: Bearer $AZURE_AI_AUTH_TOKEN' \
    -h 'content-type: application/json' \
    -d '{
            "role": "user",
            "content": "Write me a poem about flowers"
        }'

# Run thread with the agent - use both agent id and thread id
curl --request POST --url 'https://YOUR-FOUNDRY-RESOURCE-NAME.services.ai.azure.com/api/projects/YOUR-PROJECT-NAME/threads/thread_123456789/runs?api-version=v1' \
    -h 'authorization: Bearer $AZURE_AI_AUTH_TOKEN' \
    -h 'content-type: application/json' \
    --data '{
        "assistant_id": "asst_123456789"
    }'

# List the messages in the thread using thread ID
curl --request GET --url 'https://YOUR-FOUNDRY-RESOURCE-NAME.services.ai.azure.com/api/projects/YOUR-PROJECT-NAME/threads/thread_123456789/messages?api-version=v1' \
    -h 'authorization: Bearer $AZURE_AI_AUTH_TOKEN' \
    -h 'content-type: application/json'

# Delete agent once done using agent id
curl --request DELETE --url 'https://YOUR-FOUNDRY-RESOURCE-NAME.services.ai.azure.com/api/projects/YOUR-PROJECT-NAME/assistants/asst_123456789?api-version=v1' \
    -h 'authorization: Bearer $AZURE_AI_AUTH_TOKEN' \
    -h 'content-type: application/json'

Quando você estiver pronto para experimentar um agente, um agente padrão será criado para você. Para conversar com este agente:

1. No painel esquerdo, selecione Playgrounds.
2. No cartão Agents playground , selecione Let's go.
3. Adicione instruções, como "Você é um assistente de escrita útil."
4. Comece a conversar com seu agente, por exemplo: "Escreva-me um poema sobre flores".

import os
from dotenv import load_dotenv
from azure.identity import DefaultAzureCredential
from azure.ai.projects import AIProjectClient

load_dotenv()

project_client = AIProjectClient(
    endpoint=os.environ["PROJECT_ENDPOINT"],
    credential=DefaultAzureCredential(),
)

agent_name = os.environ["AGENT_NAME"]
openai_client = project_client.get_openai_client()

# Optional Step: Create a conversation to use with the agent
conversation = openai_client.conversations.create()
print(f"Created conversation (id: {conversation.id})")

# Chat with the agent to answer questions
response = openai_client.responses.create(
    conversation=conversation.id, #Optional conversation context for multi-turn
    extra_body={"agent": {"name": agent_name, "type": "agent_reference"}},
    input="What is the size of France in square miles?",
)
print(f"Response output: {response.output_text}")

# Optional Step: Ask a follow-up question in the same conversation
response = openai_client.responses.create(
    conversation=conversation.id,
    extra_body={"agent": {"name": agent_name, "type": "agent_reference"}},
    input="And what is the capital city?",
)
print(f"Response output: {response.output_text}")

using Azure.AI.Projects;
using Azure.AI.Projects.OpenAI;
using Azure.Identity;
using OpenAI.Responses;

#pragma warning disable OPENAI001

string projectEndpoint = Environment.GetEnvironmentVariable("PROJECT_ENDPOINT")
    ?? throw new InvalidOperationException("Missing environment variable 'PROJECT_ENDPOINT'");
string modelDeploymentName = Environment.GetEnvironmentVariable("MODEL_DEPLOYMENT_NAME")
    ?? throw new InvalidOperationException("Missing environment variable 'MODEL_DEPLOYMENT_NAME'");
string agentName = Environment.GetEnvironmentVariable("AGENT_NAME")
    ?? throw new InvalidOperationException("Missing environment variable 'AGENT_NAME'");

AIProjectClient projectClient = new(new Uri(projectEndpoint), new AzureCliCredential());

// Optional Step: Create a conversation to use with the agent
ProjectConversation conversation = projectClient.OpenAI.Conversations.CreateProjectConversation();

ProjectResponsesClient responsesClient = projectClient.OpenAI.GetProjectResponsesClientForAgent(
    defaultAgent: agentName,
    defaultConversationId: conversation.Id);

// Chat with the agent to answer questions
ResponseResult response = responsesClient.CreateResponse("What is the size of France in square miles?");
Console.WriteLine(response.GetOutputText());

// Optional Step: Ask a follow-up question in the same conversation
response = responsesClient.CreateResponse("And what is the capital city?");
Console.WriteLine(response.GetOutputText());

import { DefaultAzureCredential } from "@azure/identity";
import { AIProjectClient } from "@azure/ai-projects";
import "dotenv/config";

const projectEndpoint = process.env["PROJECT_ENDPOINT"] || "<project endpoint>";
const deploymentName = process.env["MODEL_DEPLOYMENT_NAME"] || "<model deployment name>";

async function main(): Promise<void> {
    const project = new AIProjectClient(projectEndpoint, new DefaultAzureCredential());
    const openAIClient = await project.getOpenAIClient();
    const response = await openAIClient.responses.create({
        model: deploymentName,
        input: "What is the size of France in square miles?",
    });
    const response2 = await openAIClient.responses.create({
        model: deploymentName,
        input: "And what is the capital city?",
        previous_response_id: response.id,
    });
    console.log(`Response output: ${response2.output_text}`);
};

main().catch(console.error);

package com.azure.ai.agents;

import com.azure.ai.agents.models.AgentReference;
import com.azure.ai.agents.models.AgentVersionDetails;
import com.azure.ai.agents.models.PromptAgentDefinition;
import com.azure.identity.AuthenticationUtil;
import com.azure.identity.DefaultAzureCredentialBuilder;
import com.openai.azure.AzureOpenAIServiceVersion;
import com.openai.azure.AzureUrlPathMode;
import com.openai.client.OpenAIClient;
import com.openai.client.okhttp.OpenAIOkHttpClient;
import com.openai.credential.BearerTokenCredential;
import com.openai.models.conversations.Conversation;
import com.openai.models.conversations.items.ItemCreateParams;
import com.openai.models.responses.EasyInputMessage;
import com.openai.models.responses.Response;
import com.openai.models.responses.ResponseCreateParams;

public class ChatWithAgent {
    public static void main(String[] args) {
        String endpoint = Configuration.getGlobalConfiguration().get("AZURE_AGENTS_ENDPOINT");
        String agentName = "MyAgent";
        
        AgentsClient agentsClient = new AgentsClientBuilder()
                .credential(new DefaultAzureCredentialBuilder().build())
                .endpoint(endpoint)
                .buildAgentsClient();

        AgentDetails agent = agentsClient.getAgent(agentName);

        Conversation conversation = conversationsClient.getConversationService().create();
        conversationsClient.getConversationService().items().create(
            ItemCreateParams.builder()
                .conversationId(conversation.id())
                .addItem(EasyInputMessage.builder()
                    .role(EasyInputMessage.Role.SYSTEM)
                    .content("You are a helpful assistant that speaks like a pirate.")
                    .build()
                ).addItem(EasyInputMessage.builder()
                    .role(EasyInputMessage.Role.USER)
                    .content("Hello, agent!")
                    .build()
            ).build()
        );

        AgentReference agentReference = new AgentReference(agent.getName()).setVersion(agent.getVersion());
        Response response = responsesClient.createWithAgentConversation(agentReference, conversation.id());

        OpenAIClient client = OpenAIOkHttpClient.builder()
            .baseUrl(endpoint.endsWith("/") ? endpoint + "openai" : endpoint + "/openai")
            .azureUrlPathMode(AzureUrlPathMode.UNIFIED)
            .credential(BearerTokenCredential.create(AuthenticationUtil.getBearerTokenSupplier(
                    new DefaultAzureCredentialBuilder().build(), "https://ai.azure.com/.default")))
            .azureServiceVersion(AzureOpenAIServiceVersion.fromString("2025-11-15-preview"))
            .build();

        ResponseCreateParams responseRequest = new ResponseCreateParams.Builder()
            .input("Hello, how can you help me?")
            .model(model)
            .build();

        Response result = client.responses().create(responseRequest);
    }
}

Substitua YOUR-FOUNDRY-RESOURCE-NAME pelos seus valores:

# Optional Step: Create a conversation to use with the agent
curl -X POST https://YOUR-FOUNDRY-RESOURCE-NAME.services.ai.azure.com/api/projects/YOUR-PROJECT-NAME/openai/conversations?api-version=2025-11-15-preview \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $AZURE_AI_AUTH_TOKEN" \
  -d '{}'
# Lets say Conversation ID created is conv_123456789. Use this in the next step

#Chat with the agent to answer questions
curl -X POST https://YOUR-FOUNDRY-RESOURCE-NAME.services.ai.azure.com/api/projects/YOUR-PROJECT-NAME/openai/responses?api-version=2025-11-15-preview \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $AZURE_AI_AUTH_TOKEN" \
  -d '{
        "agent": {"type": "agent_reference", "name": "MyAgent"},
        "conversation" : "<YOUR_CONVERSATION_ID>",
        "input" : "What is the size of France in square miles?"
    }'

#Optional Step: Ask a follow-up question in the same conversation
curl -X POST https://YOUR-FOUNDRY-RESOURCE-NAME.services.ai.azure.com/api/projects/YOUR-PROJECT-NAME/openai/responses?api-version=2025-11-15-preview \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $AZURE_AI_AUTH_TOKEN" \
  -d '{
        "agent": {"type": "agent_reference", "name": "MyAgent"},
        "conversation" : "<YOUR_CONVERSATION_ID>",
        "input" : "And what is the capital city?"
    }'

Interaja com seu agente.

1. Adicione instruções, como "Você é um assistente de escrita útil."
2. Comece a conversar com seu agente, por exemplo, "Escreva um poema sobre o sol".
3. Acompanhe com "Que tal um haiku?"

Substitua pelo endpoint o seu ponto de extremidade neste código.

from azure.ai.projects import AIProjectClient
from azure.identity import DefaultAzureCredential
from azure.ai.agents.models import ListSortOrder, FileSearchTool

project = AIProjectClient(
    endpoint="https://your-foundry-resource-name.ai.azure.com/api/projects/project-name",
    credential=DefaultAzureCredential(),
)

# Upload file and create vector store
file = project.agents.files.upload(file_path="./product_info_1.md", purpose=FilePurpose.AGENTS)
vector_store = project.agents.vector_stores.create_and_poll(file_ids=[file.id], name="my_vectorstore")

# Create file search tool and agent
file_search = FileSearchTool(vector_store_ids=[vector_store.id])
agent = project.agents.create_agent(
    model="gpt-4o",
    name="my-assistant",
    instructions="You are a helpful assistant and can search information from uploaded files",
    tools=file_search.definitions,
    tool_resources=file_search.resources,
)

# Create thread and process user message
thread = project.agents.threads.create()
project.agents.messages.create(thread_id=thread.id, role="user", content="Hello, what Contoso products do you know?")
run = project.agents.runs.create_and_process(thread_id=thread.id, agent_id=agent.id)

# Handle run status
if run.status == "failed":
    print(f"Run failed: {run.last_error}")

# Print thread messages
messages = project.agents.messages.list(thread_id=thread.id, order=ListSortOrder.ASCENDING)
for message in messages:
    if message.run_id == run.id and message.text_messages:
        print(f"{message.role}: {message.text_messages[-1].text.value}")

# Cleanup resources
project.agents.vector_stores.delete(vector_store.id)
project.agents.files.delete(file_id=file.id)
project.agents.delete_agent(agent.id)

using Azure;
using Azure.Identity;
using Azure.AI.Agents.Persistent;

// Creating the Client for agents and vector stores
var projectEndpoint = System.Environment.GetEnvironmentVariable("AZURE_AI_ENDPOINT");
var modelDeploymentName = System.Environment.GetEnvironmentVariable("AZURE_AI_MODEL");
PersistentAgentsClient client = new(projectEndpoint, new DefaultAzureCredential());

PersistentAgentFileInfo uploadedAgentFile = client.Files.UploadFile(
    filePath: "product_info_1.md",
    purpose: PersistentAgentFilePurpose.Agents);

// Create a vector store with the file and wait for it to be processed.
// If you do not specify a vector store, create_message will create a vector store with a default expiration policy of seven days after they were last active

Dictionary<string, string> fileIds = new()
{
    { uploadedAgentFile.Id, uploadedAgentFile.Filename }
};

PersistentAgentsVectorStore vectorStore = client.VectorStores.CreateVectorStore(
    name: "my_vector_store");

// Add file ID to vector store.
VectorStoreFile vctFile = client.VectorStores.CreateVectorStoreFile(
    vectorStoreId: vectorStore.Id,
    fileId: uploadedAgentFile.Id
);

Console.WriteLine($"Added file to vector store. The id file in the vector store is {vctFile.Id}.");

FileSearchToolResource fileSearchToolResource = new FileSearchToolResource();
fileSearchToolResource.VectorStoreIds.Add(vectorStore.Id);

// Create an Agent with toolResources and process Agent run
PersistentAgent agent = client.Administration.CreateAgent(
        model: modelDeploymentName,
        name: "SDK Test Agent - Retrieval",
        instructions: "You are a helpful agent that can help fetch data from files you know about.",
        tools: new List<ToolDefinition> { new FileSearchToolDefinition() },
        toolResources: new ToolResources() { FileSearch = fileSearchToolResource });

// Create thread for communication
PersistentAgentThread thread = client.Threads.CreateThread();

// Create message to thread
PersistentThreadMessage messageResponse = client.Messages.CreateMessage(
    thread.Id,
    MessageRole.User,
    "Can you give me information on how to mount the product?");

// Run the Agent
ThreadRun run = client.Runs.CreateRun(thread, agent);

// Wait for the run to complete
// This is a blocking call, so it will wait until the run is completed
do
{
    Thread.Sleep(TimeSpan.FromMilliseconds(500));
    run = client.Runs.GetRun(thread.Id, run.Id);
}
while (run.Status == RunStatus.Queued
    || run.Status == RunStatus.InProgress);

// Create a list of messages in the thread and write them to the console.
Pageable<PersistentThreadMessage> messages = client.Messages.GetMessages(
    threadId: thread.Id,
    order: ListSortOrder.Ascending
);
WriteMessages(messages, fileIds);

// Delete the thread and agent after use
client.VectorStores.DeleteVectorStore(vectorStore.Id);
client.Files.DeleteFile(uploadedAgentFile.Id);
client.Threads.DeleteThread(thread.Id);
client.Administration.DeleteAgent(agent.Id);

// Helper method to write messages to the console
static void WriteMessages(IEnumerable<PersistentThreadMessage> messages, Dictionary<string, string> fileIds)
{
    foreach (PersistentThreadMessage threadMessage in messages)
    {
        Console.Write($"{threadMessage.CreatedAt:yyyy-MM-dd HH:mm:ss} - {threadMessage.Role,10}: ");
        foreach (MessageContent contentItem in threadMessage.ContentItems)
        {
            if (contentItem is MessageTextContent textItem)
            {
                if (threadMessage.Role == MessageRole.Agent && textItem.Annotations.Count > 0)
                {
                    string strMessage = textItem.Text;
                    foreach (MessageTextAnnotation annotation in textItem.Annotations)
                    {
                        if (annotation is MessageTextFilePathAnnotation pathAnnotation)
                        {
                            strMessage = replaceReferences(fileIds, pathAnnotation.FileId, pathAnnotation.Text, strMessage);
                        }
                        else if (annotation is MessageTextFileCitationAnnotation citationAnnotation)
                        {
                            strMessage = replaceReferences(fileIds, citationAnnotation.FileId, citationAnnotation.Text, strMessage);
                        }
                    }
                    Console.Write(strMessage);
                }
                else
                {
                    Console.Write(textItem.Text);
                }
            }
            else if (contentItem is MessageImageFileContent imageFileItem)
            {
                Console.Write($"<image from ID: {imageFileItem.FileId}");
            }
            Console.WriteLine();
        }
    }
}

// Helper method to replace file references in the text
static string replaceReferences(Dictionary<string, string> fileIds, string fileID, string placeholder, string text)
{
    if (fileIds.TryGetValue(fileID, out string replacement))
        return text.Replace(placeholder, $" [{replacement}]");
    else
        return text.Replace(placeholder, $" [{fileID}]");
}

// Upload a file named product_info_1.md
console.log(`\n==================== 🕵️  FILE AGENT ====================`);
const __dirname = path.dirname(fileURLToPath(import.meta.url));
const filePath = path.join(__dirname, '../data/product_info_1.md');
const fileStream = fs.createReadStream(filePath);
fileStream.on('data', (chunk: string | Buffer) => {
    console.log(`Read ${chunk.length} bytes of data.`);
});
const file = await client.agents.files.upload(fileStream, 'assistants', {
    fileName: 'product_info_1.md'
});
console.log(`Uploaded file, ID: ${file.id}`);
const vectorStore = await client.agents.vectorStores.create({
    fileIds: [file.id], // Associate the uploaded file with the vector store
    name: 'my_vectorstore'
});
console.log('\n---------------- 🗃️ Vector Store Info ----------------');
console.table([
    {
        'Vector Store ID': vectorStore.id,
        'Usage (bytes)': vectorStore.usageBytes,
        'File Count': vectorStore.fileCounts?.total ?? 'N/A'
    }
]);

// Create an Agent and a FileSearch tool
const fileSearchTool = ToolUtility.createFileSearchTool([vectorStore.id]);
const fileAgent = await client.agents.createAgent(deployment, {
    name: 'my-file-agent',
    instructions: 'You are a helpful assistant and can search information from uploaded files',
    tools: [fileSearchTool.definition],
    toolResources: fileSearchTool.resources
});

// Create a thread and message
const fileSearchThread = await client.agents.threads.create({ toolResources: fileSearchTool.resources });
const filePrompt = 'What are the steps to setup the TrailMaster X4 Tent?';
console.log(`\n---------------- 📝 User Prompt ---------------- \n${filePrompt}`);
await client.agents.messages.create(fileSearchThread.id, 'user', filePrompt);

// Create run
let fileSearchRun = await client.agents.runs.create(fileSearchThread.id, fileAgent.id).stream();

for await (const eventMessage of fileSearchRun) {
    if (eventMessage.event === DoneEvent.Done) {
        console.log(`Run completed: ${eventMessage.data}`);
    }
    if (eventMessage.event === ErrorEvent.Error) {
        console.log(`An error occurred. ${eventMessage.data}`);
    }
}

const fileSearchMessagesIterator = await client.agents.messages.list(fileSearchThread.id);
const fileAssistantMessage = await getAssistantMessage(fileSearchMessagesIterator);
console.log(`\n---------------- 💬 Response ---------------- \n`);
printAssistantMessage(fileAssistantMessage);

// Clean up
console.log(`\n---------------- 🧹 Clean Up File Agent ----------------`);
client.agents.vectorStores.delete(vectorStore.id);
client.agents.files.delete(file.id);
client.agents.deleteAgent(fileAgent.id);
console.log(`Deleted VectorStore, File, and FileAgent. FileAgent ID: ${fileAgent.id}`);

package com.azure.ai.foundry.samples;

import java.io.IOException;
import java.nio.file.Files;
import java.nio.file.Path;

import com.azure.ai.agents.persistent.PersistentAgentsClient;
import com.azure.ai.agents.persistent.PersistentAgentsClientBuilder;
import com.azure.ai.agents.persistent.PersistentAgentsAdministrationClient;
import com.azure.ai.agents.persistent.models.CreateAgentOptions;
import com.azure.ai.agents.persistent.models.CreateThreadAndRunOptions;
import com.azure.ai.agents.persistent.models.PersistentAgent;
import com.azure.ai.agents.persistent.models.ThreadRun;
import com.azure.core.exception.HttpResponseException;
import com.azure.core.util.logging.ClientLogger;
import com.azure.identity.DefaultAzureCredentialBuilder;

/**
 * Sample demonstrating agent creation with document capabilities using Azure AI Agents Persistent SDK.
 * 
 * This sample shows how to:
 * - Set up authentication with Azure credentials
 * - Create a temporary document file for demonstration purposes
 * - Create a persistent agent with custom instructions for document search
 * - Start a thread and run with the agent that can access document content
 * - Work with file-based knowledge sources for agent interactions
 * 
 * Environment variables:
 * - AZURE_ENDPOINT: Optional fallback. The base endpoint for your Azure AI service if PROJECT_ENDPOINT is not provided.
 * - PROJECT_ENDPOINT: Required. The endpoint for your Azure AI Project.
 * - MODEL_DEPLOYMENT_NAME: Optional. The model deployment name (defaults to "gpt-4o").
 * - AGENT_NAME: Optional. The name to give to the created agent (defaults to "java-file-search-agent").
 * - AGENT_INSTRUCTIONS: Optional. The instructions for the agent (defaults to document-focused instructions).
 * 
 * Note: This sample demonstrates the creation of an agent that can process document content.
 * In a real-world scenario, you might want to integrate with Azure AI Search or similar services
 * for more advanced document processing capabilities.
 * 
 * SDK Features Demonstrated:
 * - Using the Azure AI Agents Persistent SDK (com.azure:azure-ai-agents-persistent:1.0.0-beta.2)
 * - Creating an authenticated client with DefaultAzureCredential
 * - Using the PersistentAgentsClientBuilder for client instantiation 
 * - Working with the PersistentAgentsAdministrationClient for agent management
 * - Creating temporary document files for agent access
 * - Adding document knowledge sources to agents
 * - Creating document-aware agents that can search and reference content
 * - Starting threads and runs for document-based Q&A
 * - Error handling for Azure service and file operations
 */
public class FileSearchAgentSample {
    private static final ClientLogger logger = new ClientLogger(FileSearchAgentSample.class);
    
    public static void main(String[] args) {
        // Load environment variables with proper error handling
        String endpoint = System.getenv("AZURE_ENDPOINT");
        String projectEndpoint = System.getenv("PROJECT_ENDPOINT");
        String modelName = System.getenv("MODEL_DEPLOYMENT_NAME");
        String agentName = System.getenv("AGENT_NAME");
        String instructions = System.getenv("AGENT_INSTRUCTIONS");

        // Check for required endpoint configuration
        if (projectEndpoint == null && endpoint == null) {
            String errorMessage = "Environment variables not configured. Required: either PROJECT_ENDPOINT or AZURE_ENDPOINT must be set.";
            logger.error("ERROR: {}", errorMessage);
            logger.error("Please set your environment variables or create a .env file. See README.md for details.");
            return;
        }
        
        // Set defaults for optional parameters
        if (modelName == null) {
            modelName = "gpt-4o";
            logger.info("No MODEL_DEPLOYMENT_NAME provided, using default: {}", modelName);
        }
        if (agentName == null) {
            agentName = "java-file-search-agent";
            logger.info("No AGENT_NAME provided, using default: {}", agentName);
        }
        if (instructions == null) {
            instructions = "You are a helpful assistant that can answer questions about documents.";
            logger.info("No AGENT_INSTRUCTIONS provided, using default instructions: {}", instructions);
        }

        logger.info("Building DefaultAzureCredential");
        var credential = new DefaultAzureCredentialBuilder().build();

        // Use AZURE_ENDPOINT as fallback if PROJECT_ENDPOINT not set
        String finalEndpoint = projectEndpoint != null ? projectEndpoint : endpoint;
        logger.info("Using endpoint: {}", finalEndpoint);

        try {
            // Build the general agents client with proper error handling
            logger.info("Creating PersistentAgentsClient with endpoint: {}", finalEndpoint);
            PersistentAgentsClient agentsClient = new PersistentAgentsClientBuilder()
                .endpoint(finalEndpoint)
                .credential(credential)
                .buildClient();

            // Derive the administration client
            logger.info("Getting PersistentAgentsAdministrationClient");
            PersistentAgentsAdministrationClient adminClient =
                agentsClient.getPersistentAgentsAdministrationClient();

            // Create sample document for demonstration
            Path tmpFile = createSampleDocument();
            logger.info("Created sample document at: {}", tmpFile);
            String filePreview = Files.readString(tmpFile).substring(0, 200) + "...";
            logger.info("{}", filePreview);

            // Create the agent with proper configuration
            logger.info("Creating agent with name: {}, model: {}", agentName, modelName);
            PersistentAgent agent = adminClient.createAgent(
                new CreateAgentOptions(modelName)
                    .setName(agentName)
                    .setInstructions(instructions)
            );
            logger.info("Agent ID: {}", agent.getId());
            logger.info("Agent model: {}", agent.getModel());

            // Start a thread and run on the general client
            logger.info("Creating thread and run with agent ID: {}", agent.getId());
            ThreadRun threadRun = agentsClient.createThreadAndRun(
                new CreateThreadAndRunOptions(agent.getId())
            );
            logger.info("ThreadRun ID: {}", threadRun.getThreadId());

            // Display success message
            logger.info("\nDemo completed successfully!");

        } catch (HttpResponseException e) {
            // Handle service-specific errors with detailed information
            int statusCode = e.getResponse().getStatusCode();
            logger.error("Service error {}: {}", statusCode, e.getMessage());
            logger.error("Refer to the Azure AI Agents documentation for troubleshooting information.");
            
        } catch (IOException e) {
            // Handle IO exceptions specifically for file operations
            logger.error("I/O error while creating sample document: {}", e.getMessage(), e);
            
        } catch (Exception e) {
            // Handle general exceptions
            logger.error("Error in file search agent sample: {}", e.getMessage(), e);
        }
    }

    /**
     * Creates a sample markdown document with cloud computing information.
     * 
     * This method demonstrates:
     * - Creating a temporary file that will be automatically deleted when the JVM exits
     * - Writing structured markdown content to the file
     * - Logging file creation and preview of content
     * 
     * In a real application, you might read existing files or create more complex documents.
     * You could also upload them to a document storage service for persistent access.
     * 
     * @return Path to the created temporary file
     * @throws IOException if an I/O error occurs during file creation or writing
     */
    private static Path createSampleDocument() throws IOException {
        logger.info("Creating sample document");
        String content = """
            # Cloud Computing Overview
            
            Cloud computing is the delivery of computing services over the internet, including servers, storage,
            databases, networking, software, analytics, and intelligence. Cloud services offer faster innovation,
            flexible resources, and economies of scale.
            
            ## Key Cloud Service Models
            
            1. **Infrastructure as a Service (IaaS)** - Provides virtualized computing resources
            2. **Platform as a Service (PaaS)** - Provides hardware and software tools over the internet
            3. **Software as a Service (SaaS)** - Delivers software applications over the internet
            
            ## Major Cloud Providers
            
            - Microsoft Azure
            - Amazon Web Services (AWS)
            - Google Cloud Platform (GCP)
            - IBM Cloud
            
            ## Benefits of Cloud Computing
            
            - Cost efficiency
            - Scalability
            - Reliability
            - Performance
            - Security
            """;
        
        Path tempFile = Files.createTempFile("cloud-doc", ".md");
        Files.writeString(tempFile, content);
        logger.info("Sample document created at: {}", tempFile);
        return tempFile;
    }
}

Substitua YOUR-FOUNDRY-RESOURCE-NAME e YOUR-PROJECT-NAME pelos seus valores:

#Upload the file
curl --request POST --url 'https://YOUR-FOUNDRY-RESOURCE-NAME.services.ai.azure.com/api/projects/YOUR-PROJECT-NAME/files?api-version=v1' \ 
    -h 'authorization: Bearer $AZURE_AI_AUTH_TOKEN' \ 
    -f purpose="assistant" \
    -f file="@product_info_1.md" #File object (not file name) to be uploaded.
#Lets say file ID created is assistant-123456789. Use this in the next step

# create vector store
curl --request POST --url 'https://YOUR-FOUNDRY-RESOURCE-NAME.services.ai.azure.com/api/projects/YOUR-PROJECT-NAME/vector_stores?api-version=v1' \ 
-h 'authorization: Bearer $AZURE_AI_AUTH_TOKEN' \ 
-h 'content-type: application/json' \ 
-d '{
        "name": "my_vectorstore",
        "file_ids": ["assistant-123456789"]
    }'
#Lets say Vector Store ID created is vs_123456789. Use this in the next step

# Create Agent for File Search
curl --request POST --url 'https://YOUR-FOUNDRY-RESOURCE-NAME.services.ai.azure.com/api/projects/YOUR-PROJECT-NAME/assistants?api-version=v1' \
    -h 'authorization: Bearer $AZURE_AI_AUTH_TOKEN' \ 
    -h 'content-type: application/json' \ 
    -d '{
        "model": "gpt-4o",
        "name": "my-assistant",
        "instructions": "You are a helpful assistant and can search information from uploaded files",
        "tools": [{"type": "file_search"}],
        "tool_resources": {"file_search": {"vector_store_ids": ["vs_123456789"]}}
    }'
#Lets say agent ID created is asst_123456789. Use this to run the agent

# Create thread
curl --request POST --url 'https://YOUR-FOUNDRY-RESOURCE-NAME.services.ai.azure.com/api/projects/YOUR-PROJECT-NAME/threads?api-version=v1' \
    -h 'authorization: Bearer $AZURE_AI_AUTH_TOKEN' \
    -h 'content-type: application/json'
#Lets say thread ID created is thread_123456789. Use this in the next step

# Create message using thread ID
curl --request POST --url 'https://YOUR-FOUNDRY-RESOURCE-NAME.services.ai.azure.com/api/projects/YOUR-PROJECT-NAME/threads/thread_123456789/messages?api-version=v1' \
    -h 'authorization: Bearer $AZURE_AI_AUTH_TOKEN' \
    -h 'content-type: application/json' \
    -d '{
            "role": "user",
            "content": "Hello, what Contoso products do you know?"
        }'

# Run thread with the agent - use both agent id and thread id
curl --request POST --url 'https://YOUR-FOUNDRY-RESOURCE-NAME.services.ai.azure.com/api/projects/YOUR-PROJECT-NAME/threads/thread_123456789/runs?api-version=v1' \
    -h 'authorization: Bearer $AZURE_AI_AUTH_TOKEN' \
    -h 'content-type: application/json' \
    --data '{
        "assistant_id": "asst_123456789"
    }'

# List the messages in the thread using thread ID
curl --request GET --url 'https://YOUR-FOUNDRY-RESOURCE-NAME.services.ai.azure.com/api/projects/YOUR-PROJECT-NAME/threads/thread_123456789/messages?api-version=v1' \
    -h 'authorization: Bearer $AZURE_AI_AUTH_TOKEN' \
    -h 'content-type: application/json'

# Delete agent once done using agent id
curl --request DELETE --url 'https://YOUR-FOUNDRY-RESOURCE-NAME.services.ai.azure.com/api/projects/YOUR-PROJECT-NAME/assistants/asst_123456789?api-version=v1' \
    -h 'authorization: Bearer $AZURE_AI_AUTH_TOKEN' \
    -h 'content-type: application/json'

1. No painel Configuração do agente, desloque-se para baixo, se necessário, para encontrar Conhecimento.
2. Selecione Adicionar.
3. Selecione Arquivos para carregar o arquivo product_info_1.md.
4. Selecione Selecionar arquivos locais em Adicionar arquivos.
5. Selecione Carregar e guardar.
6. Altere as instruções dos agentes, como "Você é um assistente útil e pode pesquisar informações de arquivos carregados".
7. Faça uma pergunta, como "Olá, quais produtos da Contoso você conhece?"
8. Para adicionar mais arquivos, selecione o ... no AgentVectorStore e, em seguida, selecione Gerenciar.