updates

agent.py

@@ -10,13 +10,15 @@ from langgraph.prebuilt import ToolNode, tools_condition
 
 from tools import (absolute, add, analyze_csv_file, analyze_excel_file,
                    arvix_search, audio_transcription, compound_interest,
-                   convert_temperature, divide, exponential, extract_text,
-                   factorial, floor_divide, get_current_time_in_timezone,
+                   convert_temperature, divide, exponential,
+                   extract_text_from_image, factorial, floor_divide,
+                   get_current_time_in_timezone,
+                   get_max_bird_species_count_from_video,
                    greatest_common_divisor, is_prime, least_common_multiple,
                    logarithm, modulus, multiply, percentage_calculator, power,
                    python_code_parser, reverse_sentence,
                    roman_calculator_converter, square_root, subtract,
-                   web_search, wiki_search)
+                   web_content_extract, web_search, wiki_search)
 
 # Load Constants
 load_dotenv()
@@ -32,7 +34,8 @@ tools = [
     is_prime, least_common_multiple, percentage_calculator,
     wiki_search, analyze_excel_file, arvix_search,
     audio_transcription, python_code_parser, analyze_csv_file,
-    extract_text, reverse_sentence
+    extract_text_from_image, reverse_sentence, web_content_extract,
+    get_max_bird_species_count_from_video
 ]
 
 # Load system prompt

tools.py

@@ -18,6 +18,8 @@ from langchain_community.tools import DuckDuckGoSearchRun
 from langchain_core.tools import tool
 from langchain_google_genai import ChatGoogleGenerativeAI
 from pytube import YouTube
+from langchain_tavily import TavilySearch
+from bs4 import BeautifulSoup
 
 
 @tool
@@ -228,7 +230,7 @@ def exponential(x: Union[int, float]) -> Union[float, str]:
     except Exception as e:
         return f"Error in exponential calculation: {str(e)}"
 
-
+"""
 @tool
 def web_search(query: str) -> str:
     """Performs a DuckDuckGo search for the given query and returns the results.
@@ -253,7 +255,7 @@ def web_search(query: str) -> str:
         return results
     except Exception as e:
         return f"Error performing web search: {str(e)}"
-
+"""
 
 @tool
 def roman_calculator_converter(value1: int, value2: int, oper: str) -> str:
@@ -700,7 +702,8 @@ def analyze_csv_file(file_path: str, query: str) -> str:
 # Extract Text Tool
 vision_llm = ChatGoogleGenerativeAI(model="gemini-1.5-pro",temperature=0)
 
-def extract_text(img_path: str) -> str:
+@tool
+def extract_text_from_image(img_path: str) -> str:
     """
     Extract text from an image file using a multimodal model.
 
@@ -828,3 +831,58 @@ def get_max_bird_species_count_from_video(url: str) -> Dict:
         "timestamp": f"{max_species_frame_time}s",
         "species_list": species_at_max
     }
+
+@tool
+def web_search(query: str) -> str:
+    """
+    Searches the web and returns a list of the most relevant URLs.
+    Use this FIRST for complex queries, metadata questions, or to find the right sources.
+    Then follow up with web_content_extract on the most promising URL.
+    """
+    try:
+        tavily_search = TavilySearch(
+            max_results=5,
+            topic="general",
+            search_depth="advanced",
+            include_raw_content=False,  # Just URLs and snippets
+        )
+
+        results = tavily_search.invoke(query)
+        # Format results to show URLs and brief descriptions
+        web_search_results = "Search Results:\n"
+        for i, result in enumerate(results["results"], 1):
+            web_search_results += f"{i}. {result['title']}: {result['url']}\n   {result['content'][:150]}...\n\n"
+
+        return web_search_results
+    except Exception as e:
+        return f"web_search tool error: {str(e)}"
+
+@tool
+def web_content_extract(url: str) -> str:
+    """
+    Extracts and analyzes specific content from a URL using BeautifulSoup.
+    Particularly effective for Wikipedia metadata pages, discussion pages, 
+    and structured web content.
+    Can be used after web_search to get detailed information.
+    """
+    try:
+
+        headers = {
+            "User-Agent": "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/91.0.4472.124 Safari/537.36"
+        }
+
+        response = requests.get(url, headers=headers, timeout=10)
+        response.raise_for_status()  # Raise exception for 4XX/5XX responses
+
+        soup = BeautifulSoup(response.text, 'html.parser')
+        for element in soup.select('script, style, footer, nav, header'):
+            if element:
+                element.decompose()          
+        text = soup.body.get_text(separator='\n', strip=True) if soup.body else soup.get_text(separator='\n', strip=True)
+
+        # Limit content length for response
+        return f"Content extracted from {url}:\n\n{text[:10000]}..." if len(text) > 10000 else text
+
+    except Exception as e:
+        return f"web_content_extract tool error: {str(e)}"
+