agent.py

from llama_index.llms.google_genai import GoogleGenAI
from llama_index.llms.gemini import Gemini
from llama_index.tools.arxiv import ArxivToolSpec
from llama_index.tools.wikipedia import WikipediaToolSpec
from llama_index.tools.duckduckgo import DuckDuckGoSearchToolSpec
from llama_index.core.tools import FunctionTool
from llama_index.core.agent.workflow import AgentWorkflow, ReActAgent
from llama_index.llms.lmstudio import LMStudio
from llama_index.core.agent.workflow import (
    AgentStream,
    AgentOutput
)
from gradio import ChatMessage
from llama_index.core.base.llms.types import ChatMessage as llama_index_chat_message

from tools import interpret_python_math_code, image_understanding, convert_audio_to_text, video_understanding, read_csv_file, read_xlsx_file
from gaia_system_prompt import GAIA_SYSTEM_PROMPT, CUSTOM_SYSTEM_PROMPT

import os
import asyncio

TIMEOUT=180 # Timeout for agent execution in seconds
GEMINI_API_KEY = os.getenv("GEMINI_TOKEN")
GEMINI_OPENAI_API_DIR = "https://generativelanguage.googleapis.com/v1beta/openai/"
GEMINI_MODEL_NAME = "gemini-2.0-flash"
LMSTUDIO_MODEL_NAME = "gemma-3-12B-it-qat-GGUF"
API_DIR = "http://host.docker.internal:1234/v1"  # LM Studio API URL

class FinalAgent:
    def __init__(self):
        # LLM Initialization
        # self.llm = GoogleGenAI(model=GEMINI_MODEL_NAME, api_key=GEMINI_API_KEY)
        self.llm = Gemini(model=GEMINI_MODEL_NAME, api_key=GEMINI_API_KEY)
        # self.llm = LMStudio(model_name=LMSTUDIO_MODEL_NAME, base_url=API_DIR, request_timeout=180, temperature=0.1)

        # Tool Initialization
        self.tools = [
            FunctionTool.from_defaults(
                fn=interpret_python_math_code,
                name="InterpretPythonMathCode",
                description="Interprets Python code for mathematical expressions."
            ),
            FunctionTool.from_defaults(
                fn=image_understanding,
                name="ImageUnderstanding",
                description="Analyzes an image and generates a response to a given question based on the image's content."
            ),
            FunctionTool.from_defaults(
                fn=convert_audio_to_text,
                name="ConvertAudioToText",
                description="Converts audio files to text using a speech-to-text model."
            ),
            FunctionTool.from_defaults(
                fn=video_understanding,
                name="VideoUnderstanding",
                description="Analyzes a video and generates a response to a given question based on the video's content."
            ),
            FunctionTool.from_defaults(
                fn=read_csv_file,
                name="ReadCSVFile",
                description="Reads a CSV file and returns its content as a string."
            ),
            FunctionTool.from_defaults(
                fn=read_xlsx_file,
                name="ReadXLSXFile",
                description="Reads an XLSX file and returns its content as a string."
            )
        ]
        self.tools.extend(
            ArxivToolSpec().to_tool_list()
        )
        self.tools.extend(
            WikipediaToolSpec().to_tool_list()
        )
        self.tools.extend(
            DuckDuckGoSearchToolSpec().to_tool_list()
        )

        # Print the tools for debugging
        print("Tools initialized:")
        for tool in self.tools:
            print(f"- {tool._metadata}")

        # Agent Workflow Initialization
        self.agent = AgentWorkflow.from_tools_or_functions(
            tools_or_functions=self.tools,
            llm=self.llm,
            system_prompt=CUSTOM_SYSTEM_PROMPT,
            timeout=TIMEOUT
        )

        # self.agent = ReActAgent(
        #     llm=self.llm,
        #     verbose=True,
        #     max_iterations=5,
        #     system_prompt=CUSTOM_SYSTEM_PROMPT,
        #     tools=self.tools
        # )

        print("FinalAgent initialized.")
    # async def __call__(self, question: str) -> str:
    #     # Example
    #     print(f"Agent received question: {question}")
    #     # fixed_answer = "This is a default answer."
    #     # print(f"Agent returning fixed answer: {fixed_answer}")
    #     # response = fixed_answer

    #     # Implement agent logic here
    #     response = ""
    #     # Run the agent with the question
    #     stream = await self.agent.run(question)
    #     response = stream.response.content
    #     # async for event in stream.stream_events():
    #     #         if isinstance(event, AgentStream):
    #     #              # Check if delta is empty
    #     #             if event.raw["choices"][0]["delta"] != {}:
    #     #                 response += event.raw["choices"][0]["delta"]["content"]
        
    #     print(f"Agent response: {response}")
        
    #     return response

    async def __call__(self, question: str) -> str:
        print(f"Agent received question: {question}")
        
        response_str = ""
        try:
            # Use arun for an async method.
            agent_chat_response = await self.agent.run(question)
            print(agent_chat_response)
            
            potential_response_obj = agent_chat_response.response

            if isinstance(potential_response_obj, ChatMessage):
                # If it's a ChatMessage, its .content attribute should hold the string
                print(f"DEBUG: Response object is ChatMessage. Role: {potential_response_obj.role}")
                response_str = potential_response_obj.content
                if response_str is None: # Handle cases where content might be None
                    print("DEBUG: ChatMessage content is None, defaulting to empty string.")
                    response_str = ""
            elif isinstance(potential_response_obj, str):
                # If it's already a string
                print("DEBUG: Response object is str.")
                response_str = potential_response_obj
            elif isinstance(potential_response_obj, llama_index_chat_message):
                # If it's a llama_index ChatMessage, use its .content attribute
                print(f"DEBUG: Response object is llama_index ChatMessage. Role: {potential_response_obj.role}")
                response_str = potential_response_obj.content
                if response_str is None:
                    print("DEBUG: llama_index ChatMessage content is None, defaulting to empty string.")
                    response_str = ""
            else:
                # Fallback if it's some other type
                print(f"Warning: Agent response was of unexpected type: {type(potential_response_obj)}. Converting to string.")
                response_str = str(potential_response_obj)

        except Exception as e:
            print(f"Error during agent execution with LLM {self.llm.__class__.__name__}: {e}")
            # Depending on requirements, you might want to return an error message or re-raise
            response_str = f"Agent error: {e}" 
        
        # Get the agent's final response between <final_answer> and </final_answer> tags
        if "<final_answer>" in response_str and "</final_answer>" in response_str:
            start_index = response_str.index("<final_answer>") + len("<final_answer>")
            end_index = response_str.index("</final_answer>")
            response_str = response_str[start_index:end_index].strip()
        else:
            print("Warning: No <final_answer> tags found in the response.")

        return response_str
    

# async def main():
#     # Example usage
#     agent = FinalAgent()
#     question = "How many studio albums were published by Mercedes Sosa between 2000 and 2009 (included)? You can use the latest 2022 version of english wikipedia."
#     answer = await agent(question)
#     print(f"Final answer: {answer}")

# if __name__ == "__main__":
#     asyncio.run(main())