Agents and chains?

[dixpac]

Hey this is a great project initiative.

I'm struggling to understand does this project supports agents and chains

For example can I accomplish this:

from dotenv import load_dotenv
from langchain_community.utilities import SQLDatabase
from langchain_openai import ChatOpenAI
from langchain_community.agent_toolkits import create_sql_agent

load_dotenv()


if __name__ == "__main__":
    llm = ChatOpenAI(model="gpt-3.5-turbo", temperature=0)
    db = SQLDatabase.from_uri("sqlite:///Chinook.db")

    agent_executor = create_sql_agent(llm, db=db, agent_type="openai-tools", verbose=True)

    while True:
        question = input("Ask a question (or type 'exit' to quit): ")
        if question.lower() == "exit":
            break

        try:
            response = agent_executor.invoke({"input": question })
            print(response)
        except Exception as e:
            print(e)
            continue

Thank you ✌️

[mattlindsey]

Hi @dixpac. There's an explanation of the langchainrb Langchain::Assistant class in the README and an example examples\assistant_chat.rb just like your Python example above. Just use the Langchain::Tool::Database tool instead of the weather tool.
It works with OpenAI now. GoogleGemini and others soon I think.
Any feedback would be great!

[dixpac]

Thanks