Update app.py

app.py

@@ -136,32 +136,32 @@ def should_continue(state: AgentState) -> str:
         return "reason"
     return "continue"
 
+
+
 def reasoning_node(state: AgentState) -> AgentState:
-    """Agent reasoning and tool selection"""
-    # Get Hugging Face API token from environment
+    import os
+    from langchain.schema import HumanMessage, AIMessage
+    from langchain_community.chat_models import ChatHuggingFace
+    from langchain.prompts import ChatPromptTemplate
+    from langchain_community.llms import HuggingFaceHub
+
+    # Ensure token is available
     token = os.environ.get("HF_TOKEN")
     if not token:
         raise ValueError("Hugging Face API token not found in environment variables")
-    
-    history = state["history"]
 
-    # FIX: Add a HumanMessage if the last message is not one
-    if not isinstance(history[-1], HumanMessage):
-        raise ValueError("Last message in history must be a HumanMessage")
-    
-    # Create the underlying LLM model
+    # Defensive: Ensure valid history
+    if not state["history"] or not isinstance(state["history"][-1], HumanMessage):
+        state["history"].append(HumanMessage(content="Continue."))
+
+    # Create the LLM
     llm = HuggingFaceHub(
         repo_id="HuggingFaceH4/zephyr-7b-beta",
         huggingfacehub_api_token=token,
-        model_kwargs={
-            "temperature": 0.1,
-            "max_new_tokens": 500
-        }
+        model_kwargs={"temperature": 0.1, "max_new_tokens": 500}
     )
-    
-    # Wrap the LLM in ChatHuggingFace
     chat_model = ChatHuggingFace(llm=llm)
-    
+
     # Build prompt
     prompt = ChatPromptTemplate.from_messages([
         ("system", (
@@ -175,32 +175,32 @@ def reasoning_node(state: AgentState) -> AgentState:
         )),
         *state['history']
     ])
-    
+
     chain = prompt | chat_model
     response = chain.invoke({
         "context": state['context'],
         "reasoning": state['reasoning'],
         "question": state['question']
     })
-    
-    # Parse response
+
     content = response.content
     reasoning, action, action_input = parse_agent_response(content)
-    
-    # Update state
+
     state['history'].append(AIMessage(content=content))
     state['reasoning'] += f"\nStep {state['iterations']+1}: {reasoning}"
-    
+    state['iterations'] += 1
+
     if "final answer" in action.lower():
         state['history'].append(AIMessage(content=f"FINAL ANSWER: {action_input}"))
     else:
-        state['history'].append({
+        # Save action tool call in context instead of history
+        state['context']['current_tool'] = {
             "tool": action,
-            "input": action_input,
-            "role": "action_request"
-        })
-    
+            "input": action_input
+        }
+
     return state
+
     
 
 def tool_node(state: AgentState) -> AgentState: