Update agent.py

agent.py

@@ -169,3 +169,44 @@ vector_store = SupabaseVectorStore(
     client=sup, embedding=emb,
     table_name=os.getenv("VECTORTABLE_NAME", "documents2"),
     query_name=os.getenv("VECTOR_QUERY_NAME","match_documents_langchain")
+
+def build_graph():
+    """Build the LangGraph agent using OpenAI ChatGPT only."""
+    # Initialize the OpenAI LLM
+    from langchain_openai import ChatOpenAI
+
+    llm = ChatOpenAI(
+        model="gpt-3.5-turbo",
+        temperature=0,
+        openai_api_key=os.getenv("OPENAI_API_KEY")
+    )
+    llm_with_tools = llm.bind_tools(tools)
+
+    # Retriever: try vector lookup first, else prompt LLM
+    def retriever(state: MessagesState):
+        query = state["messages"][0].content
+        hits = vector_store.similarity_search(query, k=1)
+        if hits:
+            # return the raw snippet as the single message
+            return {"messages": [sys_msg, HumanMessage(content=hits[0].page_content)]}
+        # fall back to LLM-with-tools on the user prompt
+        resp = llm_with_tools.invoke([sys_msg] + state["messages"])
+        return {"messages": [resp]}
+
+    # Assistant: always call LLM-with-tools
+    def assistant(state: MessagesState):
+        resp = llm_with_tools.invoke(state["messages"])
+        return {"messages": [resp]}
+
+    # Wire up the graph
+    builder = StateGraph(MessagesState)
+    builder.add_node("retriever", retriever)
+    builder.add_node("assistant", assistant)
+    builder.add_node("tools", ToolNode(tools))
+    builder.add_edge(START, "retriever")
+    builder.add_edge("retriever", "assistant")
+    builder.add_conditional_edges("assistant", tools_condition)
+    builder.add_edge("tools", "assistant")
+
+    return builder.compile()
+