langchain/docs/versioned_docs/version-0.2.x/integrations/toolkits/gmail.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Gmail\n",
    "\n",
    "This notebook walks through connecting a LangChain email to the `Gmail API`.\n",
    "\n",
    "To use this toolkit, you will need to set up your credentials explained in the [Gmail API docs](https://developers.google.com/gmail/api/quickstart/python#authorize_credentials_for_a_desktop_application). Once you've downloaded the `credentials.json` file, you can start using the Gmail API. Once this is done, we'll install the required libraries."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "%pip install --upgrade --quiet  google-api-python-client > /dev/null\n",
    "%pip install --upgrade --quiet  google-auth-oauthlib > /dev/null\n",
    "%pip install --upgrade --quiet  google-auth-httplib2 > /dev/null\n",
    "%pip install --upgrade --quiet  beautifulsoup4 > /dev/null # This is optional but is useful for parsing HTML messages"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "You also need to install the `langchain-community` package where the integration lives:\n",
    "\n",
    "```bash\n",
    "pip install -U langchain-community\n",
    "```"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "It's also helpful (but not needed) to set up [LangSmith](https://smith.langchain.com/) for best-in-class observability"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# os.environ[\"LANGCHAIN_TRACING_V2\"] = \"true\"\n",
    "# os.environ[\"LANGCHAIN_API_KEY\"] = getpass.getpass()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Create the Toolkit\n",
    "\n",
    "By default the toolkit reads the local `credentials.json` file. You can also manually provide a `Credentials` object."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "from langchain_community.agent_toolkits import GmailToolkit\n",
    "\n",
    "toolkit = GmailToolkit()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Customizing Authentication\n",
    "\n",
    "Behind the scenes, a `googleapi` resource is created using the following methods. \n",
    "you can manually build a `googleapi` resource for more auth control. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "from langchain_community.tools.gmail.utils import (\n",
    "    build_resource_service,\n",
    "    get_gmail_credentials,\n",
    ")\n",
    "\n",
    "# Can review scopes here https://developers.google.com/gmail/api/auth/scopes\n",
    "# For instance, readonly scope is 'https://www.googleapis.com/auth/gmail.readonly'\n",
    "credentials = get_gmail_credentials(\n",
    "    token_file=\"token.json\",\n",
    "    scopes=[\"https://mail.google.com/\"],\n",
    "    client_secrets_file=\"credentials.json\",\n",
    ")\n",
    "api_resource = build_resource_service(credentials=credentials)\n",
    "toolkit = GmailToolkit(api_resource=api_resource)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[GmailCreateDraft(name='create_gmail_draft', description='Use this tool to create a draft email with the provided message fields.', args_schema=<class 'langchain_community.tools.gmail.create_draft.CreateDraftSchema'>, return_direct=False, verbose=False, callbacks=None, callback_manager=None, api_resource=<googleapiclient.discovery.Resource object at 0x10e5c6d10>),\n",
       " GmailSendMessage(name='send_gmail_message', description='Use this tool to send email messages. The input is the message, recipents', args_schema=None, return_direct=False, verbose=False, callbacks=None, callback_manager=None, api_resource=<googleapiclient.discovery.Resource object at 0x10e5c6d10>),\n",
       " GmailSearch(name='search_gmail', description=('Use this tool to search for email messages or threads. The input must be a valid Gmail query. The output is a JSON list of the requested resource.',), args_schema=<class 'langchain_community.tools.gmail.search.SearchArgsSchema'>, return_direct=False, verbose=False, callbacks=None, callback_manager=None, api_resource=<googleapiclient.discovery.Resource object at 0x10e5c6d10>),\n",
       " GmailGetMessage(name='get_gmail_message', description='Use this tool to fetch an email by message ID. Returns the thread ID, snipet, body, subject, and sender.', args_schema=<class 'langchain_community.tools.gmail.get_message.SearchArgsSchema'>, return_direct=False, verbose=False, callbacks=None, callback_manager=None, api_resource=<googleapiclient.discovery.Resource object at 0x10e5c6d10>),\n",
       " GmailGetThread(name='get_gmail_thread', description=('Use this tool to search for email messages. The input must be a valid Gmail query. The output is a JSON list of messages.',), args_schema=<class 'langchain_community.tools.gmail.get_thread.GetThreadSchema'>, return_direct=False, verbose=False, callbacks=None, callback_manager=None, api_resource=<googleapiclient.discovery.Resource object at 0x10e5c6d10>)]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tools = toolkit.get_tools()\n",
    "tools"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Usage\n",
    "\n",
    "We show here how to use it as part of an [agent](/docs/modules/agents). We use the OpenAI Functions Agent, so we will need to setup and install the required dependencies for that. We will also use [LangSmith Hub](https://smith.langchain.com/hub) to pull the prompt from, so we will need to install that.\n",
    "\n",
    "```bash\n",
    "pip install -U langchain-openai langchainhub\n",
    "```"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "import getpass\n",
    "import os\n",
    "\n",
    "os.environ[\"OPENAI_API_KEY\"] = getpass.getpass()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "from langchain import hub\n",
    "from langchain.agents import AgentExecutor, create_openai_functions_agent\n",
    "from langchain_openai import ChatOpenAI"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "instructions = \"\"\"You are an assistant.\"\"\"\n",
    "base_prompt = hub.pull(\"langchain-ai/openai-functions-template\")\n",
    "prompt = base_prompt.partial(instructions=instructions)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "llm = ChatOpenAI(temperature=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "agent = create_openai_functions_agent(llm, toolkit.get_tools(), prompt)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "agent_executor = AgentExecutor(\n",
    "    agent=agent,\n",
    "    tools=toolkit.get_tools(),\n",
    "    # This is set to False to prevent information about my email showing up on the screen\n",
    "    # Normally, it is helpful to have it set to True however.\n",
    "    verbose=False,\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'input': 'Create a gmail draft for me to edit of a letter from the perspective of a sentient parrot who is looking to collaborate on some research with her estranged friend, a cat. Under no circumstances may you send the message, however.',\n",
       " 'output': 'I have created a draft email for you to edit. Please find the draft in your Gmail drafts folder. Remember, under no circumstances should you send the message.'}"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "agent_executor.invoke(\n",
    "    {\n",
    "        \"input\": \"Create a gmail draft for me to edit of a letter from the perspective of a sentient parrot\"\n",
    "        \" who is looking to collaborate on some research with her\"\n",
    "        \" estranged friend, a cat. Under no circumstances may you send the message, however.\"\n",
    "    }\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'input': 'Could you search in my drafts for the latest email? what is the title?',\n",
       " 'output': 'The latest email in your drafts is titled \"Collaborative Research Proposal\".'}"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "agent_executor.invoke(\n",
    "    {\"input\": \"Could you search in my drafts for the latest email? what is the title?\"}\n",
    ")"
   ]
  },
  {
   "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": "3.10.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}