AI Agents with LangChain & Autonomous Workflows · Aula

Construindo seu primeiro agente simples

Siga um guia passo a passo para configurar seu ambiente e construir um agente básico de IA usando LangChain.

Construindo seu primeiro agente simples é uma aula grátis de AI Agents with LangChain & Autonomous Workflows no CoddyKit. Esta é a aula 3 de 4. Você pode ler a aula completa abaixo gratuitamente — depois pratica ao vivo no navegador com um editor de código integrado e um tutor de IA 24/7. Faz parte do caminho de aprendizado de AI Agents with LangChain & Autonomous Workflows, e seu progresso é sincronizado entre a web e o app CoddyKit. O curso de AI Agents with LangChain & Autonomous Workflows inclui 4 aulas no total.

Partes desta aula ainda não foram traduzidas e aparecem em inglês.

Build Your First Agent!

Time to build your first AI agent with LangChain — connecting an LLM to basic tools so it can answer your queries intelligently.

Set Up Your Python Environment

First, set up Python 3.9+ and a virtual environment to isolate dependencies. Create one with venv, then activate it before you install anything.

Install LangChain & Dependencies

Now install LangChain and the OpenAI integration. Run the pip command inside your activated virtual environment.

Connect to an LLM

Your agent needs a brain. Initialize ChatOpenAI with your API key stored as the OPENAI_API_KEY environment variable — never hardcode it.

import os
from langchain_openai import ChatOpenAI

def main():
    # In a real setup, ensure OPENAI_API_KEY is set
    # For demonstration, we'll assume it's available
    # os.environ["OPENAI_API_KEY"] = "sk-..." # DON'T hardcode!

    try:
        llm = ChatOpenAI(model="gpt-3.5-turbo", temperature=0)
        print("LLM initialized successfully!")
        # You can test it:
        # response = llm.invoke("Hello, LLM!")
        # print(response.content)
    except Exception as e:
        print(f"Error initializing LLM: {e}")
        print("Please ensure OPENAI_API_KEY is set.")

if __name__ == "__main__":
    main()

Agents Need Tools

An agent is an LLM plus tools — functions or APIs that let it act on the world: search, run code, calculate. Here we give ours a calculator.

import os
from langchain_openai import ChatOpenAI
from langchain.tools import Tool
from langchain_community.utilities import LLMMathChain

def main():
    # Ensure OPENAI_API_KEY is set
    llm = ChatOpenAI(model="gpt-3.5-turbo", temperature=0)

    # Create a basic math tool
    llm_math_chain = LLMMathChain.from_llm(llm)
    math_tool = Tool.from_function(
        func=llm_math_chain.run,
        name="Calculator",
        description="Useful for when you need to answer questions about math."
    )
    tools = [math_tool]
    print("Basic math tool created and ready!")

if __name__ == "__main__":
    main()

Choosing Your Agent Type

The agent type defines how it reasons. We'll use create_react_agent — ReAct means Reason and Act: it plans, observes results, then refines.

Combining Components

Now combine the pieces: create_react_agent takes your LLM, tools, and prompt to build the logic, then AgentExecutor wraps it to make it runnable.

import os
from langchain_openai import ChatOpenAI
from langchain.tools import Tool
from langchain_community.utilities import LLMMathChain
from langchain.agents import AgentExecutor, create_react_agent
from langchain_core.prompts import ChatPromptTemplate

def main():
    llm = ChatOpenAI(model="gpt-3.5-turbo", temperature=0)
    llm_math_chain = LLMMathChain.from_llm(llm)
    math_tool = Tool.from_function(
        func=llm_math_chain.run,
        name="Calculator",
        description="Useful for when you need to answer questions about math."
    )
    tools = [math_tool]

    # Define the prompt for the agent
    prompt = ChatPromptTemplate.from_messages([
        ("system", "You are a helpful AI assistant. Use tools when necessary."),
        ("human", "{input}"),
        ("placeholder", "{agent_scratchpad}")
    ])

    # Create the agent
    agent = create_react_agent(llm, tools, prompt)

    # Create the agent executor
    agent_executor = AgentExecutor(agent=agent, tools=tools, verbose=True)
    print("Agent assembled and ready for action!")

if __name__ == "__main__":
    main()

Asking Your Agent a Question

With the agent assembled, give it a task via invoke. Ask a math question and watch it reach for its calculator tool to answer.

import os
from langchain_openai import ChatOpenAI
from langchain.tools import Tool
from langchain_community.utilities import LLMMathChain
from langchain.agents import AgentExecutor, create_react_agent
from langchain_core.prompts import ChatPromptTemplate

def main():
    llm = ChatOpenAI(model="gpt-3.5-turbo", temperature=0)
    llm_math_chain = LLMMathChain.from_llm(llm)
    math_tool = Tool.from_function(
        func=llm_math_chain.run,
        name="Calculator",
        description="Useful for when you need to answer questions about math."
    )
    tools = [math_tool]

    prompt = ChatPromptTemplate.from_messages([
        ("system", "You are a helpful AI assistant. Use tools when necessary."),
        ("human", "{input}"),
        ("placeholder", "{agent_scratchpad}")
    ])

    agent = create_react_agent(llm, tools, prompt)
    agent_executor = AgentExecutor(agent=agent, tools=tools, verbose=True)

    print("Asking the agent: 'What is 123 multiplied by 456?'")
    response = agent_executor.invoke({"input": "What is 123 multiplied by 456?"})
    print("\nAgent's final answer:")
    print(response["output"])

if __name__ == "__main__":
    main()

Understanding the Agent's Response

With verbose=True you see the agent's thought process: Thought, Action, Action Input, Observation, then Final Answer. That transparency is key to debugging.

Your Complete First Agent!

Here's your complete first agent. Set your API key, run it, and try changing the input question to see it reason through different problems.

import os
from langchain_openai import ChatOpenAI
from langchain.tools import Tool
from langchain_community.utilities import LLMMathChain
from langchain.agents import AgentExecutor, create_react_agent
from langchain_core.prompts import ChatPromptTemplate

def main():
    # 1. Set up your LLM
    # Ensure OPENAI_API_KEY is set as an environment variable
    # os.environ["OPENAI_API_KEY"] = "YOUR_KEY_HERE" # DO NOT hardcode!
    try:
        llm = ChatOpenAI(model="gpt-3.5-turbo", temperature=0)
    except Exception as e:
        print(f"Error: {e}. Please ensure OPENAI_API_KEY is set.")
        return

    # 2. Define your tools
    llm_math_chain = LLMMathChain.from_llm(llm)
    math_tool = Tool.from_function(
        func=llm_math_chain.run,
        name="Calculator",
        description="Useful for when you need to answer questions about math."
    )
    tools = [math_tool]

    # 3. Define the agent's prompt
    prompt = ChatPromptTemplate.from_messages([
        ("system", "You are a helpful AI assistant. Use tools when necessary."),
        ("human", "{input}"),
        ("placeholder", "{agent_scratchpad}")
    ])

    # 4. Create the agent
    agent = create_react_agent(llm, tools, prompt)

    # 5. Create the AgentExecutor
    agent_executor = AgentExecutor(agent=agent, tools=tools, verbose=True)

    # 6. Run the agent
    print("\n--- Running the Agent ---")
    question = "What is the square root of 144 plus 25?"
    print(f"Agent input: '{question}'")
    response = agent_executor.invoke({"input": question})
    print("\n--- Agent's Final Answer ---")
    print(response["output"])

if __name__ == "__main__":
    main()

Agent Components Check

You've seen how to build a basic agent. Which of the following are essential components when constructing a LangChain agent using create_react_agent?

Recap: Your First Agent!

Recap: you built a real agent — set up the environment, connected an LLM, added a tool, assembled it with create_react_agent and AgentExecutor, and watched it think.

Perguntas Frequentes

A aula “Construindo seu primeiro agente simples” é grátis?

Sim — o texto completo de “Construindo seu primeiro agente simples” é grátis para ler aqui na web. Para praticá-la interativamente (um editor de código integrado e um tutor de IA 24/7) e desbloquear o restante do curso de AI Agents with LangChain & Autonomous Workflows, atualize para CoddyKit PRO. O curso de AI Agents with LangChain & Autonomous Workflows inclui 4 aulas no total.

O que vou aprender em “Construindo seu primeiro agente simples”?

Siga um guia passo a passo para configurar seu ambiente e construir um agente básico de IA usando LangChain. Você pratica AI Agents with LangChain & Autonomous Workflows com código prático que executa diretamente no navegador, e um tutor de IA 24/7 responde suas dúvidas enquanto trabalha na aula.

Preciso ter experiência prévia para começar AI Agents with LangChain & Autonomous Workflows?

Nenhuma experiência prévia é necessária. AI Agents with LangChain & Autonomous Workflows no CoddyKit é estruturado para alunos iniciantes até avançados, então você pode começar aqui ou desde o início e aprender no seu ritmo. Esta é a aula 3 de 4.

Quanto tempo leva a aula “Construindo seu primeiro agente simples”?

A maioria das aulas CoddyKit leva cerca de 5–10 minutos. Cada uma é compacta e interativa, então você faz progresso constante e retoma exatamente de onde parou entre web e app.

Posso escrever e executar código nesta aula de AI Agents with LangChain & Autonomous Workflows?

Sim. Cada aula de AI Agents with LangChain & Autonomous Workflows inclui um editor de código integrado, então você escreve e executa código real direto no navegador e recebe feedback de IA instantaneamente — nenhuma configuração local necessária.

Todas as aulas deste curso

Compreendendo agentes de IA e LLMs
Explicação dos componentes principais do LangChain
Construindo seu primeiro agente simples
Dando memória e estado de conversa aos agentes

← Voltar para AI Agents with LangChain & Autonomous Workflows