Update tools.py

tools.py

@@ -31,6 +31,9 @@ Classification of questions:
 import wikipedia 
 @tool
 def wiki_search(question: str) -> str:
+    """
+    Search Wikipedia and return a short summary based on the input question.
+    """
     wikipedia.set_lang("en")
     try:
         search_results = wikipedia.search(question)
@@ -47,6 +50,9 @@ import re
 from youtube_transcript_api import YouTubeTranscriptApi 
 @tool
 def youtube_transcript(question: str) -> str:
+    """
+    Get the video fromyoutube and return a ttranscription of it.
+    """
     match = re.search(r"(?:v=|youtu\.be/)([\w\-]{11})", question)
     if not match:
         return "No YouTube video ID found."
@@ -61,6 +67,9 @@ def youtube_transcript(question: str) -> str:
 
 @tool
 def reverse_string(question: str) -> str:
+    """
+    Reverse the input question and return it.
+    """
     try:
         reversed_part = question.split('"')[1]
         return reversed_part[::-1]
@@ -70,6 +79,9 @@ def reverse_string(question: str) -> str:
 
 @tool
 def python_repl(code: str) -> str:
+    """
+    Execute a python code and return local variables.
+    """
     try:
         local_vars = {}
         exec(code, {}, local_vars)
@@ -84,6 +96,9 @@ from transformers import pipeline
 asr_pipeline = pipeline("automatic-speech-recognition", model="openai/whisper-tiny")
 @tool
 def speech_recognition(audio_path: str) -> str:
+    """
+    Extract a text from the input audio file and return a text.
+    """
     try:
         result = asr_pipeline(audio_path)
         return result["text"]
@@ -95,6 +110,9 @@ def speech_recognition(audio_path: str) -> str:
 import pandas as pd 
 @tool
 def excel_parser_tool(file_path: str) -> str:
+    """
+    Sum sales of food category and return it.
+    """
     try:
         df = pd.read_excel(file_path)  # Ensure file has a 'Category' and 'Sales' column
         food_sales = df[df["Category"] == "Food"]["Sales"].sum()
@@ -111,6 +129,9 @@ blip_processor = BlipProcessor.from_pretrained("Salesforce/blip-image-captioning
 blip_model = BlipForConditionalGeneration.from_pretrained("Salesforce/blip-image-captioning-base")
 @tool
 def chess_image_tool(image_path: str) -> str:
+    """
+    Analyze chess position on the image and return the best move.
+    """
     try:
         raw_image = Image.open(image_path).convert("RGB")
         question = "What is the best move for black in this chess position?"
@@ -128,6 +149,9 @@ import requests
 from bs4 import BeautifulSoup
 @tool
 def stat_api(question: str) -> str:
+    """
+    Analyze statistics on the link https://www.baseball-reference.com/teams/NYY/1977.shtml and return an answer to the input question.
+    """
     try:
         # Hardcoded example for the 1977 Yankees walks leader
         url = "https://www.baseball-reference.com/teams/NYY/1977.shtml"