Code Summary: Give the LLM Access to Tools

This code allows the LLM to use external tools developed previously, expanding its capabilities beyond language processing.

Link to code on GitHub

Import Packages

Update the imported packages in the main.py file with the following:

import key_param
from pymongo import MongoClient
from langchain.agents import tool
from typing import List
from langchain_openai import ChatOpenAI
from langchain_core.prompts import ChatPromptTemplate, MessagesPlaceholder
import voyageai

Build Prompt and Give Access to Tools

Update the main function to include the prompt template, binding tools, and create the pipeline.

def main():
    """
    Main function to initialize and execute the graph.
    """
    
    mongodb_client, vs_collection, full_collection = init_mongodb()
    
    tools = [
        get_information_for_question_answering,
        get_page_content_for_summarization
    ]
    
    llm = ChatOpenAI(openai_api_key=key_param.openai_api_key, temperature=0, model="gpt-4o")
    
    prompt = ChatPromptTemplate.from_messages(
        [
            (
                "You are a helpful AI assistant."
                " You are provided with tools to answer questions and summarize technical documentation related to MongoDB."
                " Think step-by-step and use these tools to get the information required to answer the user query."
                " Do not re-run tools unless absolutely necessary."
                " If you are not able to get enough information using the tools, reply with I DON'T KNOW."
                " You have access to the following tools: {tool_names}."
            ),
            MessagesPlaceholder(variable_name="messages"),
        ]
    )
    
    prompt = prompt.partial(tool_names=", ".join([tool.name for tool in tools]))
    
    bind_tools = llm.bind_tools(tools)
    
    llm_with_tools = prompt | bind_tools
    
    tool_call_check = llm_with_tools.invoke(["What are some best practices for data backups in MongoDB?"]).tool_calls
    print("Tool call check:")
    print(tool_call_check)