feat: Increase processed questions in `app.py` and use `ToolMessage` for tool call results in `agent.py`.

agent.py

@@ -6,7 +6,7 @@ from typing import TypedDict, List, Dict, Any, Optional, Union
 from langchain_core import tools
 from langgraph.graph import StateGraph, START, END
 from langchain_huggingface import ChatHuggingFace, HuggingFaceEndpoint, HuggingFacePipeline
-from langchain_core.messages import HumanMessage, AIMessage, SystemMessage
+from langchain_core.messages import HumanMessage, AIMessage, SystemMessage, ToolMessage
 from langchain_core.tools import tool
 from langchain_community.document_loaders import WikipediaLoader
 from ddgs import DDGS
@@ -374,14 +374,16 @@ def answer_message(state: AgentState) -> AgentState:
         for tool_call in tool_calls:
             name = tool_call["name"]
             args = tool_call["args"]
+            tool_call_id = tool_call.get("id")
             print(f"Calling tool: {name} with args: {args}")
             try:
                 tool = tools_by_name[name]
                 tool_result = tool.invoke(args)
             except Exception as e:
                 tool_result = f"Error executing tool {name}: {str(e)}"
-                
-            messages.append(HumanMessage(content=f"Tool result ({name}):\n{tool_result}"))
+            
+            # Using ToolMessage allows the model to map the result back perfectly to its request
+            messages.append(ToolMessage(content=str(tool_result), tool_call_id=tool_call_id, name=name))
 
     # If we exhausted all steps without an answer, force a draft response
     if draft_response is None:

app copy.py

@@ -57,7 +57,7 @@ questions_url = f"{DEFAULT_API_URL}/questions"
 response = requests.get(questions_url, timeout=15)
 response.raise_for_status()
 questions_data = response.json()
-for item in questions_data[3:4]:
+for item in questions_data[:5]:
     question_text = item.get("question")
     if question_text is None:
         continue