{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "fbaba4a9",
   "metadata": {},
   "source": [
    "# Quick Start\n",
    "\n",
    "Build a tool-calling chat widget for notebooks using LangChain. The widget is an `anywidget` + React UI, so it plays nicely with other Jupyter widgets.\n",
    "\n",
    "## 1) Install\n",
    "\n",
    "```console\n",
    "pip install langchain-widget\n",
    "```\n",
    "\n",
    "If you don't already have JupyterLab:\n",
    "\n",
    "```console\n",
    "pip install jupyterlab\n",
    "```\n",
    "\n",
    "## 2) Configure your model\n",
    "\n",
    "Set the API key for your provider. For OpenAI:\n",
    "\n",
    "```console\n",
    "export OPENAI_API_KEY=\"...\"\n",
    "```\n",
    "\n",
    "Tip: you can also use a local `.env` file and `python-dotenv` (the examples below call `load_dotenv()`):\n",
    "\n",
    "```console\n",
    "OPENAI_API_KEY=...\n",
    "```\n",
    "\n",
    "## 3) Create a widget"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "0a34ec55",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "bbbacd462f5048d4a4cdf6761a4bb158",
       "version_major": 2,
       "version_minor": 1
      },
      "text/plain": [
       "<langchain_widget.widget.LangChainWidget object at 0x7fa484187010>"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from langchain_widget import LangChainWidget\n",
    "from langchain_core.tools import tool\n",
    "from langchain_openai import ChatOpenAI\n",
    "from dotenv import load_dotenv\n",
    "\n",
    "load_dotenv()\n",
    "\n",
    "chat_model = ChatOpenAI(model=\"gpt-4o-mini\")\n",
    "\n",
    "@tool\n",
    "def add(a: int, b: int) -> int:\n",
    "    \"Add two integers.\"\n",
    "    return a + b\n",
    "\n",
    "w = LangChainWidget(\n",
    "    chat_model=chat_model,\n",
    "    tools=[add],\n",
    "    system_prompt=\"You are a helpful assistant.\",\n",
    ")\n",
    "w"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d203daa5",
   "metadata": {},
   "source": [
    "## 4) Pass context from your notebook\n",
    "\n",
    "You can inject extra context from your analysis or other widgets (selection state, current parameters, computed results, etc.)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "66291a6e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "bbbacd462f5048d4a4cdf6761a4bb158",
       "version_major": 2,
       "version_minor": 1
      },
      "text/plain": [
       "<langchain_widget.widget.LangChainWidget object at 0x7fa484187010>"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "w.add_context(title=\"Notebook context\", content=\"Selected atoms: [0, 3, 7]\")\n",
    "w"
   ]
  }
 ],
 "metadata": {
  "language_info": {
   "name": "python"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}