Update app.py

app.py

@@ -1,87 +1,69 @@
 import os
 from flask import Flask, render_template, request, jsonify
 from flask_cors import CORS
-from dotenv import load_dotenv
-
-from langchain_classic.agents import AgentExecutor, create_react_agent
-from langchain_classic.tools import Tool
-from langchain_classic.memory import ConversationBufferMemory
 from langchain_huggingface import HuggingFaceEndpoint
-from langchain_core.prompts import PromptTemplate  # NEW: Import the object type needed
-from duckduckgo_search import DDGS
+from langchain.agents import initialize_agent, Tool, AgentType
+from langchain_community.tools import DuckDuckGoSearchRun
+from langchain.memory import ConversationBufferMemory
 
-load_dotenv()
 app = Flask(__name__)
 CORS(app)
 
-HF_TOKEN = os.getenv("HF_TOKEN")
+# 1. Initialize the LLM with the correct 2026 Router settings
+# Make sure your HF Token is in your Space's 'Secrets' tab!
+sec_token = os.getenv("HUGGINGFACEHUB_API_TOKEN")
 
 llm = HuggingFaceEndpoint(
     repo_id="meta-llama/Llama-3.2-3B-Instruct",
-    huggingfacehub_api_token=HF_TOKEN,
-    temperature=0.7,
-    provider="hf-inference"
+    task="text-generation",
+    huggingfacehub_api_token=sec_token,
+    timeout=300,
+    # This prevents the 404 by using the standard inference provider
+    server_kwargs={"wait_for_model": True} 
 )
 
-def custom_ddg_search(query: str):
-    try:
-        with DDGS() as ddgs:
-            results = list(ddgs.text(query, max_results=3))
-            return "\n".join([f"{r['title']}: {r['body']}" for r in results])
-    except:
-        return "Search error. Try again."
-
+# 2. Setup Tools
+search = DuckDuckGoSearchRun()
 tools = [
     Tool(
-        name="web_search", 
-        func=custom_ddg_search, 
-        description="Useful for when you need to answer questions about current events."
+        name="web_search",
+        func=search.run,
+        description="Useful for answering questions about current events or real-time data."
     )
 ]
 
-# FIX: Manually define the ReAct prompt to avoid Hub/Client version conflicts
-template = """Answer the following questions as best you can. You have access to the following tools:
-
-{tools}
-
-Use the following format:
-
-Question: the input question you must answer
-Thought: you should always think about what to do
-Action: the action to take, should be one of [{tool_names}]
-Action Input: the input to the action
-Observation: the result of the action
-... (this Thought/Action/Action Input/Observation can repeat N times)
-Thought: I now know the final answer
-Final Answer: the final answer to the original input question
-
-Begin!
-
-Question: {input}
-Thought:{agent_scratchpad}"""
-
-prompt = PromptTemplate.from_template(template)
-
+# 3. Setup Memory & Agent
 memory = ConversationBufferMemory(memory_key="chat_history", return_messages=True)
 
-agent = create_react_agent(llm, tools, prompt)
-agent_executor = AgentExecutor(
-    agent=agent, 
-    tools=tools, 
-    memory=memory, 
+agent_executor = initialize_agent(
+    tools, 
+    llm, 
+    agent=AgentType.CONVERSATIONAL_REACT_DESCRIPTION, 
+    verbose=True, 
+    memory=memory,
     handle_parsing_errors=True
 )
 
-@app.route("/")
+@app.route('/')
 def index():
-    return render_template("index.html")
+    return render_template('index.html')
 
-@app.route("/ask", methods=["POST"])
+@app.route('/ask', methods=['POST'])
 def ask():
-    data = request.json
-    user_query = data.get("query")
-    result = agent_executor.invoke({"input": user_query})
-    return jsonify({"answer": result["output"]})
+    try:
+        data = request.get_json()
+        user_query = data.get("query")
+        
+        # Using .invoke instead of .run for better stability in 2026
+        response = agent_executor.invoke({"input": user_query})
+        
+        return jsonify({"answer": response["output"]})
+    except Exception as e:
+        print(f"Error: {str(e)}")
+        # Check if the error is a 404 to provide better feedback
+        if "404" in str(e):
+            return jsonify({"answer": "System Error: Model route not found. Please ensure you have accepted the Llama 3.2 license on Hugging Face."}), 500
+        return jsonify({"answer": f"Backend Error: {str(e)}"}), 500
 
 if __name__ == "__main__":
     app.run(host="0.0.0.0", port=7860)