Estou estudando Pythn com Langchain, com o objetivo de criar um chatbot que responde perguntas pesquisando no duckduck go ou outra ferramenta de busca. Antes de chegar no chat realmente, eu fazendo alguns testes e, a IA simplesmente não usa as ferramentas que eu disponibilizei pra ela. a IA roda, mas não faz as pesquisas para conseguir responder.

Código:

from langchain import hub
from typing import Annotated
from langgraph.prebuilt import create_react_agent
from langgraph.graph import StateGraph
from langgraph.graph.message import add_messages
from langgraph.prebuilt import ToolNode, tools_condition
from typing_extensions import TypedDict
from langchain_core.messages import AIMessage, HumanMessage, SystemMessage, AnyMessage
from langchain_community.chat_models import ChatLlamaCpp
from langchain_community.utilities import DuckDuckGoSearchAPIWrapper
from langchain_community.utilities.tavily_search import TavilySearchAPIWrapper
from langchain_community.tools.ddg_search.tool import DuckDuckGoSearchResults
from langchain_community.tools.tavily_search  import TavilySearchResults

class State(TypedDict):
    messages: Annotated[list, add_messages]


graph_builder = StateGraph(State)

model = ChatLlamaCpp(
    #model_path="E:/DEV/brincandoDeAi/brincando-de-ai-python/llm/Qwen2-0.5B-Instruct.Q8_0.gguf", #Qwen 2-0 5B Instruct Q6 GUFF
    model_path="E:/DEV/brincandoDeAi/brincando-de-ai-python/llm/Meta-Llama-3-8B-Instruct.Q6_K.gguf", #Meta Llama 3 8B Instruct Q6 GUFF
    #model_path="E:/DEV/brincandoDeAi/brincando-de-ai-python/llm/TinyLlama-1.1B-Chat-v1.0-UltraQ-Imat-NEO1-Q8_0-imat.gguf", #TinyLlama 1 1.1B Chat Q6 GUFF
    temperature=0.5,
    top_p=1,
    verbose=True,
    n_ctx=10000
)
ddgSearch = DuckDuckGoSearchAPIWrapper()
ddg_search_tool = DuckDuckGoSearchResults(verbose=True, api_wrapper=ddgSearch)
search = TavilySearchAPIWrapper(tavily_api_key="tvly-X")
search_tool = TavilySearchResults(api_wrapper=search, max_results=2)
tools = [search_tool, ddg_search_tool]


model_tools = model.bind_tools(tools)
question = "What events occurred in July 2024?"

agent = create_react_agent(model=model_tools, tools=tools, debug=True)

def print_stream(stream):
    for s in stream:
        message = s["messages"][-1]
        if isinstance(message, tuple):
            print(message)
        else:
            message.pretty_print()

response = agent.invoke(
    {
    "messages": [
    SystemMessage(content="You are a helpful assistant. You are a helpful AI. Answer the following question as best you can. In the end, show all links of the search tool if you used any."),
    SystemMessage(content='''Use a tool at least one time. Call a tool using it name, for example: 'Action: duckduckgo_results_json'. Use the exactly following format in this order:
                Question: the input question you must answer
                Thought: you should always think about what to do,
                Action: the action to take, should be one of your tools,
                Action Input: the input to the action,
                Observation: the result of the action,
                ... (this Thought/Action/Action Input/Observation can repeat N times in this exactly order),
                Thought: I now know the final answer,
                Final Answer: the final answer to the original Human input message
                Begin!
                '''),
    HumanMessage(content=question)
    ]}
)

print("------------------------------------------------------------n------------------------------------------------------------n------------------------------------------------------------")
print(response)

My response is:
I'm happy to help! However, I have to inform you that the date you provided is in the future. As of now, we don't know what will happen on July 14th, 2024.nnBut if you're looking for information on past events involving Donald Trump, I'd be happy to help with that!"

So, it could not give a answer. but, if it make a search, it can answer.
How to fix it?
Using langchain 0.2v

I am reading the docs but i can’t figure it out