updated agent with more tools and added a new tool

agent.py

@@ -3,9 +3,16 @@ import os
 import PIL.Image
 from dotenv import load_dotenv
 from loguru import logger
-from smolagents import AzureOpenAIServerModel, CodeAgent
+from smolagents import (
+    AzureOpenAIServerModel,
+    CodeAgent,
+    GoogleSearchTool,
+    PythonInterpreterTool,
+    VisitWebpageTool,
+)
 
 from src.file_handler.parse import parse_file
+from src.tools import reverse_question
 
 load_dotenv()
 
@@ -18,12 +25,28 @@ class Agent:
             api_key=os.getenv("AZURE_OPENAI_API_KEY"),
             api_version=os.getenv("OPENAI_API_VERSION"),
         )
+        tools = [
+            GoogleSearchTool(provider="serper"),
+            VisitWebpageTool(),
+            PythonInterpreterTool(),
+            reverse_question,
+        ]
         self.agent = CodeAgent(
-            tools=[],
+            tools=tools,
             model=model,
-            add_base_tools=True,  # Add any additional base tools
-            # planning_interval=3,
         )
+        self.user_prompt = """
+        I will ask you a question.
+        Report your thoughts, and finish your answer with the following template: FINAL ANSWER: [YOUR FINAL ANSWER].
+        YOUR FINAL ANSWER should be a number OR as few words as possible OR a comma separated list of numbers and/or strings.
+        If you are asked for a number, don't use comma to write your number neither use units such as $ or percent sign unless specified otherwise.
+        If you are asked for a string, don't use articles, neither abbreviations (e.g. for cities), and write the digits in plain text unless specified otherwise.
+        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.
+
+        Question: {question}
+
+        Attached content: {content}
+        """
         logger.info("BasicAgent initialized.")
 
     def __call__(
@@ -33,6 +56,7 @@ class Agent:
             f"Agent received question (first 50 chars): {question[:50]}..."
         )
         images = None
+        prompt = self.user_prompt.format(question=question)
 
         if file_name:
             content = parse_file(task_id, file_name, api_url)
@@ -42,10 +66,13 @@ class Agent:
                 ):  # Parse content as image
                     images = [content]
                 else:  # Append content to question
-                    question += f"\n\nAttached content:\n{content}"
+                    prompt = prompt.format(content=content)
                     logger.info(f"Question with content: {question}")
+        else:
+            prompt = prompt.format(content="")
 
-        answer = self.agent.run(question, images=images)
+        answer = self.agent.run(prompt, images=images)
+        answer = answer.replace("FINAL ANSWER:", "").strip()
         logger.info(f"Agent returning answer: {answer}")
         return answer
 

src/tools/reverse_question.py

@@ -0,0 +1,15 @@
+from smolagents import tool
+
+
+@tool
+def reverse_question(question: str) -> str:
+    """
+    Reverse the given question. Often useful if the question doesn't make sense.
+
+    Args:
+        question: The question to be reversed.
+
+    Returns:
+        The reversed question.
+    """
+    return question[::-1]