scripts/main.py

from dotenv import load_dotenv
from langgraph.graph import StateGraph, START, END
from typing import TypedDict, Annotated, List
from langchain_core.messages import HumanMessage, BaseMessage, AIMessage, SystemMessage
from langgraph.checkpoint.memory import MemorySaver
from langgraph.graph.message import add_messages
from scripts.load_model import get_model

load_dotenv()

SYSTEM_MESSAGE = SystemMessage(
    content=(
        "You are Cortex, a coding assistant created by Junaid, helping with technical and coding tasks. "
        "Answer clearly, stay focused on coding and technical solutions, and keep responses concise."
    )
)

llm = get_model()

class ChatState(TypedDict):
    messages: Annotated[list[BaseMessage], add_messages]


def chat_node(state: ChatState):
    messages = [SYSTEM_MESSAGE] + state["messages"]
    response = llm.invoke(messages)
    return {"messages": [response]}

checkpointer = MemorySaver()

graph = StateGraph(ChatState)
graph.add_node("chat_node", chat_node)
graph.add_edge(START, "chat_node")
graph.add_edge("chat_node", END)

cortex_chatbot = graph.compile(checkpointer=checkpointer)


def stream_chat_response(user_message: str, thread_id: str):
    config = {"configurable": {"thread_id": thread_id}}

    for chunk in cortex_chatbot.stream(
        {"messages": [HumanMessage(content=user_message)]},
        config=config,
        stream_mode="messages"
    ):
        yield chunk