experiments/ExternalAPIs/MSGraph/OAI-MSGraph-Users-Agent.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# A MS Graph API (Users) Agent\n",
    "---\n",
    "* Collaborators:\n",
    "    * Roberto Rodriguez (@Cyb3rWard0g)\n",
    "* References:\n",
    "    * https://python.langchain.com/docs/use_cases/apis\n",
    "    * https://python.langchain.com/docs/integrations/toolkits/openapi\n",
    "    * https://learn.microsoft.com/en-us/graph/api/resources/users?view=graph-rest-1.0"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Download MS Graph (User) OpenAPI specification\n",
    "\n",
    "Location: https://github.com/microsoftgraph/msgraph-sdk-powershell/blob/dev/openApiDocs/v1.0/Users.yml"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Load Open API Specification"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "import yaml\n",
    "from langchain.agents.agent_toolkits.openapi.spec import reduce_openapi_spec\n",
    "\n",
    "current_directory = os.path.abspath('')\n",
    "data_directory = os.path.join(current_directory, \"data\")\n",
    "msgraph_api_file = os.path.join(data_directory, \"OpenAPI-MSGraph-v1.0-Users.yml\")\n",
    "\n",
    "raw_msgraph_api_spec = yaml.load(open(msgraph_api_file,encoding='utf-8').read(), Loader=yaml.Loader)\n",
    "msgraph_api_spec = reduce_openapi_spec(raw_msgraph_api_spec,dereference=False)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Define LangChain Authorization Header Function"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "from azure.identity import InteractiveBrowserCredential\n",
    "from langchain.requests import RequestsWrapper\n",
    "\n",
    "def construct_auth_headers(raw_spec: dict):\n",
    "    credential = InteractiveBrowserCredential()\n",
    "    access_token = (credential.get_token(\"https://graph.microsoft.com/.default\")).token\n",
    "    return {\"Authorization\": f\"Bearer {access_token}\"}\n",
    "\n",
    "\n",
    "# Get API credentials.\n",
    "headers = construct_auth_headers(raw_msgraph_api_spec)\n",
    "requests_wrapper = RequestsWrapper(headers=headers)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## MS Graph API Spec"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "140"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "endpoints = [\n",
    "    (route, operation)\n",
    "    for route, operations in raw_msgraph_api_spec[\"paths\"].items()\n",
    "    for operation in operations\n",
    "    if operation in [\"get\", \"post\"]\n",
    "]\n",
    "len(endpoints)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Initialize Open AI Key"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "import os\n",
    "from dotenv import load_dotenv\n",
    "import openai\n",
    "\n",
    "# Get your key: https://platform.openai.com/account/api-keys\n",
    "load_dotenv()\n",
    "openai.api_key = os.getenv(\"OPENAI_API_KEY\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Define OpenAI Chat API"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "from langchain.chat_models import ChatOpenAI\n",
    "from langchain.agents.agent_toolkits.openapi import planner\n",
    "\n",
    "llm = ChatOpenAI(model_name=\"gpt-4\", temperature=0.0)\n",
    "msgraph_agent = planner.create_openapi_agent(msgraph_api_spec, requests_wrapper, llm)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Run Query"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "\n",
      "\u001b[1m> Entering new AgentExecutor chain...\u001b[0m\n",
      "\u001b[32;1m\u001b[1;3mAction: api_planner\n",
      "Action Input: Search for user with id 3b83a100-f4b1-4b9a-934e-5626c7c00dbd\u001b[0m\n",
      "Observation: \u001b[36;1m\u001b[1;3m1. GET /users/{user-id} to retrieve the properties and relationships of the user with the id 3b83a100-f4b1-4b9a-934e-5626c7c00dbd.\u001b[0m\n",
      "Thought:\u001b[32;1m\u001b[1;3mI'm ready to execute the API call.\n",
      "Action: api_controller\n",
      "Action Input: GET /users/{user-id} to retrieve the properties and relationships of the user with the id 3b83a100-f4b1-4b9a-934e-5626c7c00dbd.\u001b[0m\n",
      "\n",
      "\u001b[1m> Entering new AgentExecutor chain...\u001b[0m\n",
      "\u001b[32;1m\u001b[1;3mI need to use the requests_get tool to make a GET request to the /users/{user-id} endpoint of the API. The user-id is 3b83a100-f4b1-4b9a-934e-5626c7c00dbd. I don't need any parameters for this request. I will extract the id and name from the response.\n",
      "\n",
      "Action: requests_get\n",
      "Action Input: \n",
      "{\n",
      "  \"url\": \"https://graph.microsoft.com/v1.0/users/3b83a100-f4b1-4b9a-934e-5626c7c00dbd\",\n",
      "  \"params\": {},\n",
      "  \"output_instructions\": {\n",
      "    \"id\": \"id\",\n",
      "    \"name\": \"displayName\"\n",
      "  }\n",
      "}\n",
      "\u001b[0m\n",
      "Observation: \u001b[36;1m\u001b[1;3m{'id': '3b83a100-f4b1-4b9a-934e-5626c7c00dbd', 'name': 'Cyb3rWard0g'}\u001b[0m\n",
      "Thought:\u001b[32;1m\u001b[1;3mI have successfully retrieved the user's id and name. The user's id is 3b83a100-f4b1-4b9a-934e-5626c7c00dbd and the name is Cyb3rWard0g. I have completed the plan.\n",
      "\n",
      "Final Answer: {'id': '3b83a100-f4b1-4b9a-934e-5626c00dbd', 'name': 'Cyb3rWard0g'}\u001b[0m\n",
      "\n",
      "\u001b[1m> Finished chain.\u001b[0m\n",
      "\n",
      "Observation: \u001b[33;1m\u001b[1;3m{'id': '3b83a100-f4b1-4b9a-934e-5626c00dbd', 'name': 'Cyb3rWard0g'}\u001b[0m\n",
      "Thought:\u001b[32;1m\u001b[1;3mI am finished executing the plan and have the information the user asked for.\n",
      "Final Answer: The user with id 3b83a100-f4b1-4b9a-934e-5626c7c00dbd is named 'Cyb3rWard0g'.\u001b[0m\n",
      "\n",
      "\u001b[1m> Finished chain.\u001b[0m\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "\"The user with id 3b83a100-f4b1-4b9a-934e-5626c7c00dbd is named 'Cyb3rWard0g'.\""
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "user_query = (\n",
    "    \"Search for user with id 3b83a100-f4b1-4b9a-934e-5626c7c00dbd\"\n",
    ")\n",
    "msgraph_agent.run(user_query)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## One Cell Execution"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "\n",
      "\u001b[1m> Entering new AgentExecutor chain...\u001b[0m\n",
      "\u001b[32;1m\u001b[1;3mAction: api_planner\n",
      "Action Input: Search for user with id 3b83a100-f4b1-4b9a-934e-5626c7c00dbd\u001b[0m\n",
      "Observation: \u001b[36;1m\u001b[1;3m1. GET /users/{user-id} to retrieve the properties and relationships of the user with the id 3b83a100-f4b1-4b9a-934e-5626c7c00dbd.\u001b[0m\n",
      "Thought:\u001b[32;1m\u001b[1;3mI'm ready to execute the API call.\n",
      "Action: api_controller\n",
      "Action Input: GET /users/{user-id} to retrieve the properties and relationships of the user with the id 3b83a100-f4b1-4b9a-934e-5626c7c00dbd.\u001b[0m\n",
      "\n",
      "\u001b[1m> Entering new AgentExecutor chain...\u001b[0m\n",
      "\u001b[32;1m\u001b[1;3mI need to use the requests_get tool to make a GET request to the /users/{user-id} endpoint of the API. The user-id in the path should be replaced with the actual user id.\n",
      "\n",
      "Action: requests_get\n",
      "Action Input: \n",
      "{\n",
      "  \"url\": \"https://graph.microsoft.com/v1.0/users/3b83a100-f4b1-4b9a-934e-5626c7c00dbd\",\n",
      "  \"params\": {},\n",
      "  \"output_instructions\": \"Extract the id and name of the user\"\n",
      "}\u001b[0m\n",
      "Observation: \u001b[36;1m\u001b[1;3mThe user's ID is \"3b83a100-f4b1-4b9a-934e-5626c7c00dbd\" and the name is \"Cyb3rWard0g\".\u001b[0m\n",
      "Thought:\u001b[32;1m\u001b[1;3mI have successfully retrieved the user's information. The plan has been executed successfully.\n",
      "Final Answer: The user's ID is \"3b83a100-f4b1-4b9a-934e-5626c7c00dbd\" and the name is \"Cyb3rWard0g\".\u001b[0m\n",
      "\n",
      "\u001b[1m> Finished chain.\u001b[0m\n",
      "\n",
      "Observation: \u001b[33;1m\u001b[1;3mThe user's ID is \"3b83a100-f4b1-4b9a-934e-5626c7c00dbd\" and the name is \"Cyb3rWard0g\".\u001b[0m\n",
      "Thought:\u001b[32;1m\u001b[1;3mI am finished executing the plan and have the information the user asked for.\n",
      "Final Answer: The user's ID is \"3b83a100-f4b1-4b9a-934e-5626c7c00dbd\" and the name is \"Cyb3rWard0g\".\u001b[0m\n",
      "\n",
      "\u001b[1m> Finished chain.\u001b[0m\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "'The user\\'s ID is \"3b83a100-f4b1-4b9a-934e-5626c7c00dbd\" and the name is \"Cyb3rWard0g\".'"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Download MS Graph API OpenAPI Specification:\n",
    "#  https://github.com/microsoftgraph/msgraph-sdk-powershell/blob/dev/openApiDocs/v1.0/Users.yml\n",
    "import os\n",
    "import yaml\n",
    "from langchain.agents.agent_toolkits.openapi.spec import reduce_openapi_spec\n",
    "\n",
    "# Read File Locally\n",
    "current_directory = os.path.abspath('')\n",
    "msgraph_api_file = os.path.join('data', \"OpenAPI-MSGraph-v1.0-Users.yml\")\n",
    "raw_msgraph_api_spec = yaml.load(open(msgraph_api_file,encoding='utf-8').read(), Loader=yaml.Loader)\n",
    "\n",
    "# Read as OpenAPI Specification\n",
    "msgraph_api_spec = reduce_openapi_spec(raw_msgraph_api_spec,dereference=False)\n",
    "\n",
    "# Define LangChain Authorization Header Function \n",
    "from azure.identity import InteractiveBrowserCredential\n",
    "from langchain.requests import RequestsWrapper\n",
    "\n",
    "def construct_auth_headers(raw_spec: dict):\n",
    "    credential = InteractiveBrowserCredential()\n",
    "    access_token = (credential.get_token(\"https://graph.microsoft.com/.default\")).token\n",
    "    return {\"Authorization\": f\"Bearer {access_token}\"}\n",
    "\n",
    "# Get Access Token -> Authorization Header.\n",
    "headers = construct_auth_headers(raw_msgraph_api_spec)\n",
    "requests_wrapper = RequestsWrapper(headers=headers)\n",
    "\n",
    "# Define OpenAI Chat API\n",
    "from langchain.chat_models import ChatOpenAI\n",
    "from langchain.agents.agent_toolkits.openapi import planner\n",
    "from dotenv import load_dotenv\n",
    "import openai\n",
    "# Initialize OpenAI Key. Get your key: https://platform.openai.com/account/api-keys\n",
    "load_dotenv()\n",
    "openai.api_key = os.getenv(\"OPENAI_API_KEY\")\n",
    "llm = ChatOpenAI(model_name=\"gpt-4\", temperature=0.0)\n",
    "\n",
    "# Create OpenAPI Agent\n",
    "msgraph_agent = planner.create_openapi_agent(msgraph_api_spec, requests_wrapper, llm)\n",
    "\n",
    "# Pass prompt (Natural Language) to Agent -> Call MS Graph API\n",
    "msgraph_agent.run((\"Search for user with id 3b83a100-f4b1-4b9a-934e-5626c7c00dbd\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.11.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}