How to add agent's tools call response to chat history

[ox1234]

What i building: I build a log analysis agent, which can call the tools to get some different log.
Question: How can i add the agent's tool call response to the chat history, i found the chat history only contains the user and agent's conversation. I also want to add the tool call response(the log raw content) to the chat history. How can i do that?

Here is my graph:

[dosubot[bot]]

Hey @ox1234! 👋 I'm here to help you with any bugs, questions, or contributions you have in mind. Let's squash those bugs together!

I couldn't find specific information on how to add an agent's tool call response to the chat history in Langflow. However, I can provide some insights based on the general structure of Langflow's chat history management.

Langflow uses the ChatHistory class to store and update chat history, which includes methods like add_message to append messages to the history. Tool responses can be integrated into the chat history by creating a ChatResponse or FileResponse object and using the add_message method to include these responses in the chat history ^[1].

For example, when a tool response is generated, you can encapsulate it in a FileResponse object and add it to the chat history. Here's a code snippet that demonstrates this process:

def update(self):
    if self.chat_cache.current_client_id in self.active_connections:
        self.last_cached_object_dict = self.chat_cache.get_last()
        # Add a new ChatResponse with the data
        chat_response = FileResponse(
            message=None,
            type="file",
            data=self.last_cached_object_dict["obj"],
            data_type=self.last_cached_object_dict["type"],
        )

        self.chat_history.add_message(self.chat_cache.current_client_id, chat_response)

This code shows how a FileResponse is created and added to the chat history when a cached object is updated. You can adapt this approach to include your tool call responses in the chat history.

To continue talking to Dosu, mention @dosu.

---

^{Help Dosu learn! Give it feedback: Great Response | Irrelevant Answer | Incorrect Sources | Too Verbose | Hallucination | Other | Bug Report}

[ox1234]

Actually i don't know how to get the tool call's repsonse, if i can, i can add it to the chat response

[dosubot[bot]]

To retrieve the tool call's response and add it to the chat history, you can use the message_response method from the ToolCallingAgentComponent class. This method executes the agent and retrieves the response, which is then returned as a Message object. The response is stored in response.data.get("text"), which contains the tool call's response.

Here's an example of how you can use this method:

async def test_tool_calling_agent_component():
    tools = [CalculatorToolComponent().build_tool()]  # Use the Calculator component as a tool
    input_value = "What is 2 + 2?"
    chat_history = []
    api_key = os.environ["OPENAI_API_KEY"]
    temperature = 0.1

    # Default OpenAI Model Component
    llm_component = OpenAIModelComponent().set(
        api_key=api_key,
        temperature=temperature,
    )
    llm = llm_component.build_model()

    agent = ToolCallingAgentComponent()
    agent.set(llm=llm, tools=[tools], chat_history=chat_history, input_value=input_value)

    # Chat output
    response = await agent.message_response()
    result_string = response.data.get("text")
    assert "4" in result_string

This result_string contains the tool call's response, which you can then add to the chat history ^[1][2].

To continue talking to Dosu, mention @dosu.

---

^{Help Dosu learn! Give it feedback: Great Response | Irrelevant Answer | Incorrect Sources | Too Verbose | Hallucination | Other | Bug Report}