Update app.py

app.py

@@ -5,12 +5,12 @@ import inspect
 import pandas as pd
 import re
 import time
+import base64
 #from tavily import TavilyClient
 from langchain_tavily import TavilySearch
 from langgraph.prebuilt import create_react_agent
 from langgraph.graph.message import add_messages
 from langgraph_supervisor import create_supervisor
-#from langchain_huggingface import HuggingFaceEndpoint, ChatHuggingFace
 from langchain_google_genai import ChatGoogleGenerativeAI
 from langchain_core.tools import tool
 from langchain_core.messages import HumanMessage, SystemMessage, BaseMessage
@@ -51,6 +51,7 @@ prompt_recomendado = """You are a general AI assistant. I will ask you a questio
                         If you are asked for a comma separated list, apply the above rules depending of whether the element to be put in the list is a number or a string.
                         To assist in your task, you can supervise other agents who perform specific tasks that could not be handled by tools, since they require the processing of another LLM. Below, I will inform you about your assistants:
                         - web_research_agent. Assign web research related tasks to this agent, prioritizing the use of Wikipedia sources
+                        - chess_position_review_agent. Assign chess position review related tasks to this agent
                         Assign work to one agent at a time, do not call agents in parallel.
                         Priorize the use of tools and another agents to help in reasoning.
                         When a file or URL is entered at the prompt, use it in tools or other agents, both are prepared to handle files and URLs."""
@@ -62,19 +63,165 @@ prompt_search = """You are a web research agent.
                    After you're done with your tasks, respond to the supervisor directly
                    Respond ONLY with the results of your work, do NOT include ANY other text."""
 
+prompt_chess = """You are a chess position reviewing agent.
+                   INSTRUCTIONS:
+                   Assist ONLY with tasks related to chess position reviewing, DO NOT do any math
+                   After you're done with your tasks, respond to the supervisor directly
+                   Respond ONLY with the results of your work, do NOT include ANY other text."""
+
 #TOOLS
 web_search = TavilySearch(
         max_results=5,
         topic="general",
     )
 
+def chess_image_to_fen_tool(task_id:str, current_player: Literal["black", "white"]) -> Dict[str,str]:
+    """
+        Convert chess image to FEN (Forsyth-Edwards Notation) notation.
+        Args:
+            image_path: Path to the image file.
+            current_player: Whose turn it is to play. Must be either 'black' or 'white'.
+        Returns:
+            JSON with FEN (Forsyth-Edwards Notation) string representing the current board position.
+    """
+    print(f"Image to Fen invocada com os seguintes parametros:")
+    print(f"task_id: {task_id}")
+    print(f"current_player: {current_player}")
+
+
+    if current_player not in ["black", "white"]:
+        raise ValueError("current_player must be 'black' or 'white'")
+
+    base64_image = download_file_as_base64(task_id)
+    if not base64_image:
+        raise ValueError("Failed to encode image to base64.")
+    base64_image_encoded =  f"data:image/jpeg;base64,{base64_image}"
+    url = CHESSVISION_TO_FEN_URL
+    payload = {
+        "board_orientation": "predict",
+        "cropped": False,
+        "current_player": "black",
+        "image": base64_image_encoded,
+        "predict_turn": False
+    }
+
+    response = requests.post(url, json=payload)
+    if response.status_code == 200:
+        dados = response.json()
+        if dados.get("success"):
+            print(f"Retorno Chessvision {dados}")
+            fen = dados.get("result")
+            fen = fen.replace("_", " ") #retorna _ no lugar de espaço em branco
+            return json.dumps({"fen": fen})
+        else:
+            raise Exception("Requisição feita, mas falhou na predição.")
+    else:
+        raise Exception(f"Erro na requisição: {response.status_code}")
+
+def chess_fen_get_best_next_move_tool(fen: str, current_player: Literal["black", "white"]) -> str:
+    """
+        Return the best move in algebraic notation.
+        Args:
+            fen: FEN (Forsyth-Edwards Notation) notation.
+        Returns:
+            Best move in algebraic notation.
+    """
+    if not fen:
+        raise ValueError("fen must be provided.")
+    if current_player not in ["black", "white"]:
+        raise ValueError("current_player must be 'black' or 'white'")
+
+
+    url = CHESS_MOVE_API
+    payload = {
+       "fen": fen,
+       "depth": 1
+    }
+
+    print(f"Buscando melhor jogada em {CHESS_MOVE_API} - {payload}")
+
+    response = requests.post(url, json=payload)
+    if response.status_code == 200:
+        #print(f"Retorno melhor jogada --> {response.text}")
+        dados = response.json()
+        move_algebric_notation = dados.get("san")
+        move = dados.get("text")
+        print(f"Melhor jogada segundo chess-api.com -> {move}")
+
+        return move_algebric_notation
+
+    else:
+        raise Exception(f"Erro na requisição: {response.status_code}")
+
+def download_file(task_id: str):
+    if not fen:
+        raise ValueError("task_id must be provided.")
+    # Construct the URL
+    url = f"https://agents-course-unit4-scoring.hf.space/files/{task_id}"
+    
+    # Send the request to download the file
+    response = requests.get(url)
+    
+    if response.status_code == 200:
+        # Get the content type from the response headers to determine the file extension
+        content_type = response.headers.get('Content-Type', '')
+        file_extension = ''
+        
+        # Map content types to file extensions
+        if 'pdf' in content_type:
+            file_extension = '.pdf'
+        elif 'jpg' in content_type or 'jpeg' in content_type:
+            file_extension = '.jpg'
+        elif 'png' in content_type:
+            file_extension = '.png'
+        elif 'txt' in content_type:
+            file_extension = '.txt'
+        elif 'zip' in content_type:
+            file_extension = '.zip'
+        elif 'mp3' in content_type:
+            file_extension = '.mp3'
+        # Add more file types as necessary
+
+        # If the extension can't be determined, default to .bin
+        if not file_extension:
+            file_extension = '.bin'
+        
+        # Set the path to the Downloads folder (adjust 'YourUsername' to your actual username)
+        save_path = os.path.join(os.path.expanduser('~'), 'Downloads', f"{task_id}_file{file_extension}")
+        
+        # Save the file with the appropriate extension
+        with open(save_path, 'wb') as f:
+            f.write(response.content)
+        print(f"File successfully downloaded and saved as {save_path}")
+
+        return save_path
+    else:
+        print(f"Failed to download the file. Status code: {response.status_code}")
+
+def download_file_as_base64(task_id: str) -> str:
+    # Construct the URL
+    url = f"https://agents-course-unit4-scoring.hf.space/files/{task_id}"
+    
+    # Send the request to download the file
+    response = requests.get(url)
+    
+    if response.status_code == 200:
+        # Encode the content to Base64
+        encoded_bytes = base64.b64encode(response.content)
+        encoded_str = encoded_bytes.decode('utf-8')  # Convert bytes to string
+        return encoded_str
+    else:
+        raise Exception(f"Failed to download the file. Status code: {response.status_code}")
+        
+
+
 tools = [web_search]
 
 #LLMS
 # Create LLM class 
 gemini_llm = ChatGoogleGenerativeAI(
     model= "gemini-2.0-flash-exp", # replace with "gemini-2.0-flash"
-    temperature=1.0,
+    temperature=0.0,
     max_tokens=None,
     timeout=None,
     max_retries=2,
@@ -91,6 +238,13 @@ web_research_agent = create_react_agent(
     name="web_research_agent"
 )
 
+chess_position_review_agent = create_react_agent(
+    model=gemini_llm,
+    tools=[chess_image_to_fen_tool,chess_fen_get_best_next_move_tool],
+    prompt=prompt_chess,
+    name="chess_position_review_agent"
+)
+
 supervisor = create_supervisor(
     model=gemini_llm,
     agents=[web_research_agent],
@@ -118,7 +272,7 @@ class BasicAgent:
 #        tools = [search_web_tool]
 #        chat_with_tools = chat.bind_tools(tools)
         
-    def __call__(self, question: str) -> str:
+    def __call__(self, question: str, path: str) -> str:
         print(f"Agent received question : {question}...")
 
         messages = {
@@ -218,7 +372,7 @@ def run_and_submit_all( profile: gr.OAuthProfile | None):
             print(f"Skipping item with missing task_id or question: {item}")
             continue
         try:
-            submitted_answer = agent(question_text)
+            submitted_answer = agent(question_text,task_id)
             answers_payload.append({"task_id": task_id, "submitted_answer": submitted_answer})
             results_log.append({"Task ID": task_id, "Question": question_text, "Submitted Answer": submitted_answer})
         except Exception as e: