Add ntbks

README.md

@@ -0,0 +1,74 @@
+# Task mAIstro
+
+Task mAIstro allows you to track your ToDos through natural conversation: 
+
+* Tell it about yourself 
+* Tell it tasks you need to complete
+* Update tasks easily through conversation
+* Ask task mAIstro what to to next 
+
+Task mAIstro combines an agent with long-term memory to produce a highly personalized task management experience. All information can be stored locally using the LangGraph Studio Desktop App or it can be deployed to LangGraph Platform.
+
+## Key Features
+
+### Natural Language Interface
+
+- Create and update tasks through natural conversation
+- No need to learn specific commands or syntax
+- Just chat with the assistant as you would with a human
+
+### Smart Memory System
+
+Task mAIstro maintains three types of memory:
+
+1. **ToDo List**
+   - Tasks with descriptions
+   - Estimated completion times
+   - Deadlines
+   - Actionable solutions
+   - Status tracking
+
+2. **User Profile**
+   - Remembers personal details
+   - Tracks preferences
+   - Maintains context about your life and work
+
+3. **Update Instructions**
+   - Learns how you prefer tasks to be managed
+   - Adapts to your organizational style
+   - Maintains your preferences for task updates
+
+## Quickstart 
+
+You can deploy the app locally with the LangGraph Studio Desktop App or to LangGraph Cloud. 
+
+### Locally 
+
+Populate the `.env` file with your `OPENAI_API_KEY` key: 
+```
+cp .env.example .env
+```
+
+Download the LangGraph Studio desktop app for Mac [here](https://github.com/langchain-ai/langgraph-studio?tab=readme-ov-file#download).
+
+Load this repository as a project in LangGraph Studio.
+
+Start chatting with the task mAIstro!
+
+### LangGraph Cloud
+
+In your LangSmith account, create a new deployment with this repository's `main` branch.
+
+Set your `OPENAI_API_KEY` as a secret when creating the deployment.
+
+Interact with the deployment through the LangGraph Studio web UI.
+
+Use the `ntbk/connecting_to_graph.ipynb` notebook to interact with the deployed graph via .
+
+## Audio UX
+
+See the `ntbk/audio_ux.ipynb` notebook for an example of how to add an audio interface to your graph. 
+
+## Learning More
+
+See [Module 5 of our LangChain Academy Course on LangGraph](https://academy.langchain.com/courses/intro-to-langgraph) to learn how to build this app from scratch! 

ntbk/connecting_to_graph.ipynb

@@ -0,0 +1,112 @@
+{
+ "cells": [
+  {
+   "cell_type": "markdown",
+   "id": "b1e3eaa4",
+   "metadata": {},
+   "source": [
+    "# Connecting To The Deployment\n",
+    "\n",
+    "Connected to the [LangSmith](https://smith.langchain.com/) deployment of `task_maistro`.\n",
+    "\n",
+    "Ensure the `LANGCHAIN_API_KEY` for the LangSmith account with your deployment is set.\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "f4de2831",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import os, getpass\n",
+    "\n",
+    "def _set_env(var: str):\n",
+    "    # Check if the variable is set in the OS environment\n",
+    "    env_value = os.environ.get(var)\n",
+    "    if not env_value:\n",
+    "        # If not set, prompt the user for input\n",
+    "        env_value = getpass.getpass(f\"{var}: \")\n",
+    "    \n",
+    "    # Set the environment variable for the current process\n",
+    "    os.environ[var] = env_value\n",
+    "\n",
+    "_set_env(\"LANGCHAIN_API_KEY\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "id": "386677ba",
+   "metadata": {},
+   "source": [
+    "### Graph connection\n",
+    "\n",
+    "Let's first test that we can connect to the graph. \n",
+    "\n",
+    "We can find the URL of the graph in the [LangSmith](https://smith.langchain.com/) UI.\n",
+    "\n",
+    "We get the `url` from the deployment page. "
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 29,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from langgraph.pregel.remote import RemoteGraph\n",
+    "from langchain_core.messages import convert_to_messages\n",
+    "from langchain_core.messages import HumanMessage, SystemMessage\n",
+    "\n",
+    "# Add your deployment URL here\n",
+    "url = \"https://task-maistro-1b681add7a2b549499bb0cd21a7e5be4.default.us.langgraph.app\"\n",
+    "# Graph name is from langgraph.json\n",
+    "graph_name = \"task_maistro\" \n",
+    "# Connect to the graph\n",
+    "remote_graph = RemoteGraph(graph_name, url=url)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 30,
+   "id": "fc8102cc",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "================================\u001b[1m Human Message \u001b[0m=================================\n",
+      "\n",
+      "Hi I'm Lance. I live in San Francisco with my wife and have a 1 year old.\n",
+      "==================================\u001b[1m Ai Message \u001b[0m==================================\n",
+      "\n",
+      "Hi Lance! It's great to meet you. I already have that information in my memory. How can I assist you today?\n"
+     ]
+    }
+   ],
+   "source": [
+    "# Interact with the graph\n",
+    "user_input = \"Hi I'm Lance. I live in San Francisco with my wife and have a 1 year old.\"\n",
+    "config = {\"configurable\": {\"user_id\": \"Lance\"}}\n",
+    "for chunk in remote_graph.stream({\"messages\": [HumanMessage(content=user_input)]}, stream_mode=\"values\", config=config):\n",
+    "    convert_to_messages(chunk[\"messages\"])[-1].pretty_print()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "d5851ecb",
+   "metadata": {},
+   "outputs": [],
+   "source": []
+  }
+ ],
+ "metadata": {
+  "language_info": {
+   "name": "python"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}

requirements.txt

@@ -2,4 +2,5 @@ langgraph
 langchain-core
 langchain-community
 langchain-openai
-trustcall
+trustcall
+ipython