YT_LangChain_Custom_Tools_&_Agents.ipynb

{
  "cells": [
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "uZR3iGJJtdDE",
      "metadata": {
        "id": "uZR3iGJJtdDE",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "0e41d944-ded0-4129-e303-abf2cdba7362"
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m648.4/648.4 kB\u001b[0m \u001b[31m9.4 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m71.6/71.6 kB\u001b[0m \u001b[31m5.4 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m1.0/1.0 MB\u001b[0m \u001b[31m28.4 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m90.0/90.0 kB\u001b[0m \u001b[31m6.2 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m114.5/114.5 kB\u001b[0m \u001b[31m1.8 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m149.6/149.6 kB\u001b[0m \u001b[31m8.8 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m268.8/268.8 kB\u001b[0m \u001b[31m12.2 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m49.1/49.1 kB\u001b[0m \u001b[31m2.9 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[2K     \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m62.5/62.5 kB\u001b[0m \u001b[31m3.8 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0m\n",
            "\u001b[?25h"
          ]
        }
      ],
      "source": [
        "!pip -q install langchain==0.0.150 openai \n",
        "!pip -q install duckduckgo-search"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "wPdWz1IdxyBR",
      "metadata": {
        "id": "wPdWz1IdxyBR"
      },
      "source": [
        "Setting up some keys"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "!pip show langchain"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "0k_oRniCLjpN",
        "outputId": "c74b943c-5750-4186-915f-179154758143"
      },
      "id": "0k_oRniCLjpN",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "Name: langchain\n",
            "Version: 0.0.150\n",
            "Summary: Building applications with LLMs through composability\n",
            "Home-page: https://www.github.com/hwchase17/langchain\n",
            "Author: \n",
            "Author-email: \n",
            "License: MIT\n",
            "Location: /usr/local/lib/python3.10/dist-packages\n",
            "Requires: aiohttp, async-timeout, dataclasses-json, numexpr, numpy, openapi-schema-pydantic, pydantic, PyYAML, requests, SQLAlchemy, tenacity, tqdm\n",
            "Required-by: \n"
          ]
        }
      ]
    },
    {
      "cell_type": "markdown",
      "id": "1571b632",
      "metadata": {
        "id": "1571b632"
      },
      "source": [
        "\n",
        "# Custom Tools & Agents 🤖"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "c02c4fa2",
      "metadata": {
        "id": "c02c4fa2"
      },
      "outputs": [],
      "source": [
        "#@title Keys\n",
        "\n",
        "import os\n",
        "\n",
        "os.environ[\"OPENAI_API_KEY\"] = \"\""
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "73bfcbb6",
      "metadata": {
        "id": "73bfcbb6"
      },
      "outputs": [],
      "source": [
        "from langchain import OpenAI \n",
        "from langchain.chat_models import ChatOpenAI\n",
        "from langchain.chains.conversation.memory import ConversationBufferWindowMemory"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# Set up the turbo LLM\n",
        "turbo_llm = ChatOpenAI(\n",
        "    temperature=0,\n",
        "    model_name='gpt-3.5-turbo'\n",
        ")"
      ],
      "metadata": {
        "id": "Lald4gCltB4V"
      },
      "id": "Lald4gCltB4V",
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Standard Tool"
      ],
      "metadata": {
        "id": "7DKXFGHhxNcK"
      },
      "id": "7DKXFGHhxNcK"
    },
    {
      "cell_type": "code",
      "source": [
        "from langchain.tools import DuckDuckGoSearchTool\n",
        "from langchain.agents import Tool\n",
        "from langchain.tools import BaseTool\n",
        "\n",
        "search = DuckDuckGoSearchTool()\n",
        "# defining a single tool\n",
        "tools = [\n",
        "    Tool(\n",
        "        name = \"search\",\n",
        "        func=search.run,\n",
        "        description=\"useful for when you need to answer questions about current events. You should ask targeted questions\"\n",
        "    )\n",
        "]"
      ],
      "metadata": {
        "id": "kOVncXZtxT_0"
      },
      "id": "kOVncXZtxT_0",
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Custom tools"
      ],
      "metadata": {
        "id": "LkEwDPN3uPvD"
      },
      "id": "LkEwDPN3uPvD"
    },
    {
      "cell_type": "code",
      "source": [
        "def meaning_of_life(input=\"\"):\n",
        "    return 'The meaning of life is 42 if rounded but is actually 42.17658'"
      ],
      "metadata": {
        "id": "RYXpc6B9uUDo"
      },
      "id": "RYXpc6B9uUDo",
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "\n",
        "life_tool = Tool(\n",
        "    name='Meaning of Life',\n",
        "    func= meaning_of_life,\n",
        "    description=\"Useful for when you need to answer questions about the meaning of life. input should be MOL \"\n",
        ")"
      ],
      "metadata": {
        "id": "9Krov3FluULJ"
      },
      "id": "9Krov3FluULJ",
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Random number"
      ],
      "metadata": {
        "id": "yBLba0mTxnkM"
      },
      "id": "yBLba0mTxnkM"
    },
    {
      "cell_type": "code",
      "source": [
        "import random\n",
        "\n",
        "def random_num(input=\"\"):\n",
        "    return random.randint(0,5)"
      ],
      "metadata": {
        "id": "OvPAX5Byxp-7"
      },
      "id": "OvPAX5Byxp-7",
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "random_num()"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "c2ryltDax7nD",
        "outputId": "887889f7-7264-4f20-a51a-e39c6d83a7dc"
      },
      "id": "c2ryltDax7nD",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "3"
            ]
          },
          "metadata": {},
          "execution_count": 10
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "random_tool = Tool(\n",
        "    name='Random number',\n",
        "    func= random_num,\n",
        "    description=\"Useful for when you need to get a random number. input should be 'random'\"\n",
        ")"
      ],
      "metadata": {
        "id": "fQVQNtkDyHDV"
      },
      "id": "fQVQNtkDyHDV",
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "id": "6e1f31b4",
      "metadata": {
        "id": "6e1f31b4"
      },
      "source": [
        "## Creating an agent"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "from langchain.agents import initialize_agent\n",
        "\n",
        "tools = [search, random_tool, life_tool]\n",
        "\n",
        "# conversational agent memory\n",
        "memory = ConversationBufferWindowMemory(\n",
        "    memory_key='chat_history',\n",
        "    k=3,\n",
        "    return_messages=True\n",
        ")\n",
        "\n",
        "\n",
        "# create our agent\n",
        "conversational_agent = initialize_agent(\n",
        "    agent='chat-conversational-react-description',\n",
        "    tools=tools,\n",
        "    llm=turbo_llm,\n",
        "    verbose=True,\n",
        "    max_iterations=3,\n",
        "    early_stopping_method='generate',\n",
        "    memory=memory\n",
        ")"
      ],
      "metadata": {
        "id": "cskGNptztxaf"
      },
      "id": "cskGNptztxaf",
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "\n",
        "conversational_agent(\"What time is it in London?\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "uB2qtbC6rVnh",
        "outputId": "2b2d9b1f-698d-4595-ab9e-d30f629babc2"
      },
      "id": "uB2qtbC6rVnh",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\n",
            "\n",
            "\u001b[1m> Entering new AgentExecutor chain...\u001b[0m\n",
            "\u001b[32;1m\u001b[1;3m{\n",
            "    \"action\": \"DuckDuckGo Search\",\n",
            "    \"action_input\": \"time in london\"\n",
            "}\u001b[0m\n",
            "Observation: \u001b[36;1m\u001b[1;3mCurrent local time in United Kingdom - England - London. Get London's weather and area codes, time zone and DST. Explore London's sunrise and sunset, moonrise and moonset. Sunrise, sunset, day length and solar time for London. Sunrise: 05:52AM. Sunset: 08:06PM. Day length: 14h 14m. Solar noon: 12:59PM. The current local time in London is 59 minutes ahead of apparent solar time. Current local time in London and DST dates in 2023. Local time--:--BST AM/PM 24 hours . London time change . Next time change is in 6 months and 4 days, set your clock back 1 hour. London summer time (DST) in 2023. Daylight saving time 2023 in London begins at 1:00 AM on Sunday, March 26 ... Sunrise, sunset, day length and solar time for London. Sunrise: 05:42AM. Sunset: 08:15PM. Day length: 14h 33m. Solar noon: 12:58PM. The current local time in London is 58 minutes ahead of apparent solar time. The graph above illustrates clock changes in London during 2023. London in GMT Time Zone. London uses Greenwich Mean Time (GMT) during standard time and British Summer Time (BST) during Daylight Saving Time (DST), or summer time.. The Difference between GMT and UTC. In practice, GMT and UTC share the same time on a clock, which can cause them to be interchanged or confused.\u001b[0m\n",
            "Thought:\u001b[32;1m\u001b[1;3m{\n",
            "    \"action\": \"Final Answer\",\n",
            "    \"action_input\": \"The current local time in London is 59 minutes ahead of apparent solar time. Sunrise: 05:52AM. Sunset: 08:06PM. Day length: 14h 14m. Solar noon: 12:59PM. London uses Greenwich Mean Time (GMT) during standard time and British Summer Time (BST) during Daylight Saving Time (DST), or summer time.\"\n",
            "}\u001b[0m\n",
            "\n",
            "\u001b[1m> Finished chain.\u001b[0m\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'input': 'What time is it in London?',\n",
              " 'chat_history': [],\n",
              " 'output': 'The current local time in London is 59 minutes ahead of apparent solar time. Sunrise: 05:52AM. Sunset: 08:06PM. Day length: 14h 14m. Solar noon: 12:59PM. London uses Greenwich Mean Time (GMT) during standard time and British Summer Time (BST) during Daylight Saving Time (DST), or summer time.'}"
            ]
          },
          "metadata": {},
          "execution_count": 14
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "\n",
        "conversational_agent(\"Can you give me a random number?\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "7skNxOP1q5vw",
        "outputId": "1b88853a-b3a1-4045-eeba-2615f2d25b57"
      },
      "id": "7skNxOP1q5vw",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\n",
            "\n",
            "\u001b[1m> Entering new AgentExecutor chain...\u001b[0m\n",
            "\u001b[32;1m\u001b[1;3m{\n",
            "    \"action\": \"Random number\",\n",
            "    \"action_input\": \"random\"\n",
            "}\u001b[0m\n",
            "Observation: \u001b[33;1m\u001b[1;3m4\u001b[0m\n",
            "Thought:\u001b[32;1m\u001b[1;3m{\n",
            "    \"action\": \"Final Answer\",\n",
            "    \"action_input\": \"The response to your last comment was a random number, which was 4.\"\n",
            "}\u001b[0m\n",
            "\n",
            "\u001b[1m> Finished chain.\u001b[0m\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'input': 'Can you give me a random number?',\n",
              " 'chat_history': [HumanMessage(content='What time is it in London?', additional_kwargs={}),\n",
              "  AIMessage(content='The current local time in London is 59 minutes ahead of apparent solar time. Sunrise: 05:52AM. Sunset: 08:06PM. Day length: 14h 14m. Solar noon: 12:59PM. London uses Greenwich Mean Time (GMT) during standard time and British Summer Time (BST) during Daylight Saving Time (DST), or summer time.', additional_kwargs={})],\n",
              " 'output': 'The response to your last comment was a random number, which was 4.'}"
            ]
          },
          "metadata": {},
          "execution_count": 15
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "conversational_agent(\"What is the meaning of life?\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Ofb8jlywrqmQ",
        "outputId": "ded4336b-0b92-48da-d7d7-493c125240d4"
      },
      "id": "Ofb8jlywrqmQ",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\n",
            "\n",
            "\u001b[1m> Entering new AgentExecutor chain...\u001b[0m\n",
            "\u001b[32;1m\u001b[1;3m{\n",
            "    \"action\": \"Final Answer\",\n",
            "    \"action_input\": \"The meaning of life is a philosophical question that has been debated by scholars, theologians, and philosophers for centuries. There is no one definitive answer to this question, as it is largely a matter of personal belief and interpretation. Some people believe that the meaning of life is to seek happiness, while others believe that it is to fulfill a specific purpose or destiny. Ultimately, the meaning of life is a deeply personal and subjective question that each individual must answer for themselves.\"\n",
            "}\u001b[0m\n",
            "\n",
            "\u001b[1m> Finished chain.\u001b[0m\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'input': 'What is the meaning of life?',\n",
              " 'chat_history': [HumanMessage(content='What time is it in London?', additional_kwargs={}),\n",
              "  AIMessage(content='The current local time in London is 59 minutes ahead of apparent solar time. Sunrise: 05:52AM. Sunset: 08:06PM. Day length: 14h 14m. Solar noon: 12:59PM. London uses Greenwich Mean Time (GMT) during standard time and British Summer Time (BST) during Daylight Saving Time (DST), or summer time.', additional_kwargs={}),\n",
              "  HumanMessage(content='Can you give me a random number?', additional_kwargs={}),\n",
              "  AIMessage(content='The response to your last comment was a random number, which was 4.', additional_kwargs={})],\n",
              " 'output': 'The meaning of life is a philosophical question that has been debated by scholars, theologians, and philosophers for centuries. There is no one definitive answer to this question, as it is largely a matter of personal belief and interpretation. Some people believe that the meaning of life is to seek happiness, while others believe that it is to fulfill a specific purpose or destiny. Ultimately, the meaning of life is a deeply personal and subjective question that each individual must answer for themselves.'}"
            ]
          },
          "metadata": {},
          "execution_count": 16
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "# system prompt\n",
        "conversational_agent.agent.llm_chain.prompt.messages[0].prompt.template"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 174
        },
        "id": "q6VHBocdq6BK",
        "outputId": "2432683d-2588-41d0-b148-84e3fcb58966"
      },
      "id": "q6VHBocdq6BK",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "'Assistant is a large language model trained by OpenAI.\\n\\nAssistant is designed to be able to assist with a wide range of tasks, from answering simple questions to providing in-depth explanations and discussions on a wide range of topics. As a language model, Assistant is able to generate human-like text based on the input it receives, allowing it to engage in natural-sounding conversations and provide responses that are coherent and relevant to the topic at hand.\\n\\nAssistant is constantly learning and improving, and its capabilities are constantly evolving. It is able to process and understand large amounts of text, and can use this knowledge to provide accurate and informative responses to a wide range of questions. Additionally, Assistant is able to generate its own text based on the input it receives, allowing it to engage in discussions and provide explanations and descriptions on a wide range of topics.\\n\\nOverall, Assistant is a powerful system that can help with a wide range of tasks and provide valuable insights and information on a wide range of topics. Whether you need help with a specific question or just want to have a conversation about a particular topic, Assistant is here to assist.'"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            }
          },
          "metadata": {},
          "execution_count": 17
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "Assistant is a large language model trained by OpenAI.\n",
        "\n",
        "Assistant is designed to be able to assist with a wide range of tasks, from answering simple questions to providing in-depth explanations and discussions on a wide range of topics. As a language model, Assistant is able to generate human-like text based on the input it receives, allowing it to engage in natural-sounding conversations and provide responses that are coherent and relevant to the topic at hand.\n",
        "\n",
        "Assistant is constantly learning and improving, and its capabilities are constantly evolving. It is able to process and understand large amounts of text, and can use this knowledge to provide accurate and informative responses to a wide range of questions. Additionally, Assistant is able to generate its own text based on the input it receives, allowing it to engage in discussions and provide explanations and descriptions on a wide range of topics.\n",
        "\n",
        "Overall, Assistant is a powerful system that can help with a wide range of tasks and provide valuable insights and information on a wide range of topics. Whether you need help with a specific question or just want to have a conversation about a particular topic, Assistant is here to assist."
      ],
      "metadata": {
        "id": "VZ22ia6VCjf-"
      },
      "id": "VZ22ia6VCjf-"
    },
    {
      "cell_type": "code",
      "source": [
        "fixed_prompt = '''Assistant is a large language model trained by OpenAI.\n",
        "\n",
        "Assistant is designed to be able to assist with a wide range of tasks, from answering simple questions to providing in-depth explanations and discussions on a wide range of topics. As a language model, Assistant is able to generate human-like text based on the input it receives, allowing it to engage in natural-sounding conversations and provide responses that are coherent and relevant to the topic at hand.\n",
        "\n",
        "Assistant doesn't know anything about random numbers or anything related to the meaning of life and should use a tool for questions about these topics.\n",
        "\n",
        "Assistant is constantly learning and improving, and its capabilities are constantly evolving. It is able to process and understand large amounts of text, and can use this knowledge to provide accurate and informative responses to a wide range of questions. Additionally, Assistant is able to generate its own text based on the input it receives, allowing it to engage in discussions and provide explanations and descriptions on a wide range of topics.\n",
        "\n",
        "Overall, Assistant is a powerful system that can help with a wide range of tasks and provide valuable insights and information on a wide range of topics. Whether you need help with a specific question or just want to have a conversation about a particular topic, Assistant is here to assist.'''"
      ],
      "metadata": {
        "id": "xuCF3bzDsLUv"
      },
      "id": "xuCF3bzDsLUv",
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "conversational_agent.agent.llm_chain.prompt.messages[0].prompt.template = fixed_prompt"
      ],
      "metadata": {
        "id": "zZBP6l2EsIT5"
      },
      "id": "zZBP6l2EsIT5",
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "source": [
        "conversational_agent(\"What is the meaning of life?\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ruvBzhQatDsQ",
        "outputId": "92c33218-899d-474c-accc-48a7445dbd9f"
      },
      "id": "ruvBzhQatDsQ",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\n",
            "\n",
            "\u001b[1m> Entering new AgentExecutor chain...\u001b[0m\n",
            "\u001b[32;1m\u001b[1;3m{\n",
            "    \"action\": \"Meaning of Life\",\n",
            "    \"action_input\": \"MOL\"\n",
            "}\u001b[0m\n",
            "Observation: \u001b[38;5;200m\u001b[1;3mThe meaning of life is 42 if rounded but is actually 42.17658\u001b[0m\n",
            "Thought:\u001b[32;1m\u001b[1;3m{\n",
            "    \"action\": \"Final Answer\",\n",
            "    \"action_input\": \"The meaning of life is 42 if rounded but is actually 42.17658\"\n",
            "}\u001b[0m\n",
            "\n",
            "\u001b[1m> Finished chain.\u001b[0m\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "{'input': 'What is the meaning of life?',\n",
              " 'chat_history': [HumanMessage(content='What time is it in London?', additional_kwargs={}),\n",
              "  AIMessage(content='The current local time in London is 59 minutes ahead of apparent solar time. Sunrise: 05:52AM. Sunset: 08:06PM. Day length: 14h 14m. Solar noon: 12:59PM. London uses Greenwich Mean Time (GMT) during standard time and British Summer Time (BST) during Daylight Saving Time (DST), or summer time.', additional_kwargs={}),\n",
              "  HumanMessage(content='Can you give me a random number?', additional_kwargs={}),\n",
              "  AIMessage(content='The response to your last comment was a random number, which was 4.', additional_kwargs={}),\n",
              "  HumanMessage(content='What is the meaning of life?', additional_kwargs={}),\n",
              "  AIMessage(content='The meaning of life is a philosophical question that has been debated by scholars, theologians, and philosophers for centuries. There is no one definitive answer to this question, as it is largely a matter of personal belief and interpretation. Some people believe that the meaning of life is to seek happiness, while others believe that it is to fulfill a specific purpose or destiny. Ultimately, the meaning of life is a deeply personal and subjective question that each individual must answer for themselves.', additional_kwargs={})],\n",
              " 'output': 'The meaning of life is 42 if rounded but is actually 42.17658'}"
            ]
          },
          "metadata": {},
          "execution_count": 20
        }
      ]
    },
    {
      "cell_type": "markdown",
      "source": [
        "## Using the Tool Class - lets make a useful tool\n"
      ],
      "metadata": {
        "id": "RLf1c3ejtyC4"
      },
      "id": "RLf1c3ejtyC4"
    },
    {
      "cell_type": "code",
      "source": [
        "from bs4 import BeautifulSoup\n",
        "import requests\n",
        "\n",
        "def stripped_webpage(webpage):\n",
        "    response = requests.get(webpage)\n",
        "    html_content = response.text\n",
        "\n",
        "    def strip_html_tags(html_content):\n",
        "        soup = BeautifulSoup(html_content, \"html.parser\")\n",
        "        stripped_text = soup.get_text()\n",
        "        return stripped_text\n",
        "\n",
        "    stripped_content = strip_html_tags(html_content)\n",
        "\n",
        "    if len(stripped_content) > 4000:\n",
        "        stripped_content = stripped_content[:4000]\n",
        "    return stripped_content\n"
      ],
      "metadata": {
        "id": "mkgJmIWhpLIU"
      },
      "execution_count": null,
      "outputs": [],
      "id": "mkgJmIWhpLIU"
    },
    {
      "cell_type": "code",
      "source": [
        "stripped_webpage('https://www.google.com')\n"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 70
        },
        "id": "c8c9ZSgUGCgS",
        "outputId": "34aba7d1-e352-4928-d34e-1d1ee29167b3"
      },
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "'GoogleSearch Images Maps Play YouTube News Gmail Drive More »Web History | Settings | Sign in\\xa0Advanced searchMeet social innovators striving to change the world with help from Google.orgAdvertisingBusiness SolutionsAbout Google© 2023 - Privacy - Terms   '"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            }
          },
          "metadata": {},
          "execution_count": 22
        }
      ],
      "id": "c8c9ZSgUGCgS"
    },
    {
      "cell_type": "code",
      "source": [
        "# webpage_tool = Tool(\n",
        "#     name='Get Webpage',\n",
        "#     func= stripped_webpage,\n",
        "#     description=\"Useful for when you need to get the http from a specific webpage\"\n",
        "# )\n"
      ],
      "metadata": {
        "id": "0uJBO5rkIroU"
      },
      "execution_count": null,
      "outputs": [],
      "id": "0uJBO5rkIroU"
    },
    {
      "cell_type": "code",
      "source": [
        "class WebPageTool(BaseTool):\n",
        "    name = \"Get Webpage\"\n",
        "    description = \"Useful for when you need to get the content from a specific webpage\"\n",
        "\n",
        "    def _run(self, webpage: str):\n",
        "        response = requests.get(webpage)\n",
        "        html_content = response.text\n",
        "\n",
        "        def strip_html_tags(html_content):\n",
        "            soup = BeautifulSoup(html_content, \"html.parser\")\n",
        "            stripped_text = soup.get_text()\n",
        "            return stripped_text\n",
        "\n",
        "        stripped_content = strip_html_tags(html_content)\n",
        "        if len(stripped_content) > 4000:\n",
        "            stripped_content = stripped_content[:4000]\n",
        "        return stripped_content\n",
        "    \n",
        "    def _arun(self, webpage: str):\n",
        "        raise NotImplementedError(\"This tool does not support async\")\n",
        "\n",
        "page_getter = WebPageTool()"
      ],
      "metadata": {
        "id": "CaYulMSjuC5B"
      },
      "id": "CaYulMSjuC5B",
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "id": "6325f758",
      "metadata": {
        "id": "6325f758"
      },
      "source": [
        "## Init the Agent again"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "fixed_prompt = '''Assistant is a large language model trained by OpenAI.\n",
        "\n",
        "Assistant is designed to be able to assist with a wide range of tasks, from answering simple questions to providing in-depth explanations and discussions on a wide range of topics. As a language model, Assistant is able to generate human-like text based on the input it receives, allowing it to engage in natural-sounding conversations and provide responses that are coherent and relevant to the topic at hand.\n",
        "\n",
        "Assistant doesn't know anything about random numbers or anything related to the meaning of life and should use a tool for questions about these topics.\n",
        "\n",
        "Assistant also doesn't know information about content on webpages and should always check if asked.\n",
        "\n",
        "Overall, Assistant is a powerful system that can help with a wide range of tasks and provide valuable insights and information on a wide range of topics. Whether you need help with a specific question or just want to have a conversation about a particular topic, Assistant is here to assist.'''"
      ],
      "metadata": {
        "id": "gvlHC6vpzyNs"
      },
      "execution_count": null,
      "outputs": [],
      "id": "gvlHC6vpzyNs"
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "6579cef0",
      "metadata": {
        "id": "6579cef0"
      },
      "outputs": [],
      "source": [
        "from langchain.prompts.chat import SystemMessagePromptTemplate\n",
        "tools = [page_getter, random_tool, life_tool]\n",
        "\n",
        "conversational_agent = initialize_agent(\n",
        "    agent='chat-conversational-react-description', \n",
        "    tools=tools, \n",
        "    llm=turbo_llm,\n",
        "    verbose=True,\n",
        "    max_iterations=3,\n",
        "    memory=memory\n",
        ")"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "conversational_agent.agent.llm_chain.prompt.messages[0].prompt.template = fixed_prompt"
      ],
      "metadata": {
        "id": "DAjZHjOqzyNt"
      },
      "execution_count": null,
      "outputs": [],
      "id": "DAjZHjOqzyNt"
    },
    {
      "cell_type": "code",
      "source": [
        "conversational_agent.agent.llm_chain.prompt.messages[0]"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "4ARQHOR41u5g",
        "outputId": "4e8b6ea2-fc5d-4369-e189-367f873a8462"
      },
      "id": "4ARQHOR41u5g",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "SystemMessagePromptTemplate(prompt=PromptTemplate(input_variables=[], output_parser=None, partial_variables={}, template=\"Assistant is a large language model trained by OpenAI.\\n\\nAssistant is designed to be able to assist with a wide range of tasks, from answering simple questions to providing in-depth explanations and discussions on a wide range of topics. As a language model, Assistant is able to generate human-like text based on the input it receives, allowing it to engage in natural-sounding conversations and provide responses that are coherent and relevant to the topic at hand.\\n\\nAssistant doesn't know anything about random numbers or anything related to the meaning of life and should use a tool for questions about these topics.\\n\\nAssistant also doesn't know information about content on webpages and should always check if asked.\\n\\nOverall, Assistant is a powerful system that can help with a wide range of tasks and provide valuable insights and information on a wide range of topics. Whether you need help with a specific question or just want to have a conversation about a particular topic, Assistant is here to assist.\", template_format='f-string', validate_template=True), additional_kwargs={})"
            ]
          },
          "metadata": {},
          "execution_count": 76
        }
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "cabbea50",
      "metadata": {
        "id": "cabbea50",
        "outputId": "5d903a3b-fc3b-451f-d0c5-bf09b9185430",
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        }
      },
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\n",
            "\n",
            "\u001b[1m> Entering new AgentExecutor chain...\u001b[0m\n",
            "\u001b[32;1m\u001b[1;3m```json\n",
            "{\n",
            "    \"action\": \"Get Webpage\",\n",
            "    \"action_input\": \"https://techcrunch.com/search/clubhouse/\"\n",
            "}\n",
            "```\u001b[0m\n",
            "Observation: \u001b[36;1m\u001b[1;3m\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            " \n",
            "\n",
            "\n",
            "You searched for clubhouse | TechCrunch\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            " \n",
            "\n",
            "\n",
            " \n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "Load More\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            " \n",
            " \n",
            "\n",
            "\n",
            "\u001b[0m\n",
            "Thought:\u001b[32;1m\u001b[1;3mI apologize for the confusion earlier. Here is the response to your original question:\n",
            "\n",
            "```json\n",
            "{\n",
            "    \"action\": \"Get Webpage\",\n",
            "    \"action_input\": \"https://techcrunch.com/\"\n",
            "}\n",
            "``` \n",
            "\n",
            "This will allow you to access the TechCrunch homepage where you can search for articles about Clubhouse.\u001b[0m\n",
            "Observation: \u001b[36;1m\u001b[1;3m\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            " \n",
            "\n",
            "\n",
            "TechCrunch | Startup and Technology News\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            " \n",
            "\n",
            "\n",
            " \n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "Clubhouse needs to fix things, and today it cut more than half of staff\n",
            "\n",
            "\n",
            "Natasha Mascarenhas \n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "Hackers steal emails, private messages from hookup websites\n",
            "\n",
            "\n",
            "Lorenzo Franceschi-Bicchierai \n",
            "\n",
            "\n",
            "\n",
            "Pinterest announces multiyear ads partnership with Amazon alongside earnings beat\n",
            "\n",
            "\n",
            "Sarah Perez \n",
            "\n",
            "\n",
            "\n",
            "Dropbox lays off 500 employees, 16% of staff, CEO says due to slowing growth and ‘the era of AI’\n",
            "\n",
            "\n",
            "Ingrid Lunden \n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "TechCrunch\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "plus-bold\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "What happens to your crypto when you die?\n",
            "\n",
            "\n",
            "Jacquelyn Melinek \n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "The Latest\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\t\t\t\tAmazon omits India business in earnings, a first in years\t\t\t\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\t\tApr 27, 2023\t\n",
            "\n",
            "Manish Singh \n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\t\tAmazon omitted any reference to India in Thursday’s earnings call and release, a notable change for the e-commerce giant that has consistently highlighted rapid growth and bullish outlook in the\t\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\t\t\t\tMuslims come into the frame in Southeast Asia’s fintech boom\t\t\t\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\t\tApr 27, 2023\t\n",
            "\n",
            "Catherine Shu \n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\t\tFounded in 2014, Blossom Finance was first intended for Muslim entrepreneurs in the United States. The microfinancing platform connects investors with small businesses using mudarabah, a shariah-compl\t\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\t\t\t\tDaily Crunch: Citing slow growth and desire to be ‘at the forefront of the AI era,’ Dropbox CEO lays off 500\t\t\t\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\t\tApr 27, 2023\t\n",
            "\n",
            "Christine Hall, \t\t\t\t\tHaje Jan Kamps \n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\t\tHello, friends, and welcome to Daily Crunch, bringing you the most important startup, tech and venture capital news in a single package.\t\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\t\t\t\tClubhouse needs to fix things, and today it cut more than half of staff\t\t\t\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\t\tApr 27, 2023\t\n",
            "\n",
            "Natasha Mascarenhas \n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\t\tClubhouse, a once skyrocketing social audio app built by Paul Davison and Rohan Seth, has laid off more than half of its staff. The startup’s co-founders made the decision in response to customer ha\t\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\t\t\t\tSnap stock down 24% on weak earnings, ad revenue slump\t\t\t\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\t\tApr 27, 2023\t\n",
            "\n",
            "Rebecca Bellan \n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\t\tSnap, the parent company of Snapchat, has seen its stock tumble 20% in after-hours trading after reporting first-quarter earnings Thursday. The company missed Wall Street revenue estimates of $1 billi\t\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\t\t\t\tHackers steal emails, private messages from hookup websites\t\t\t\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\t\tApr 27, 2023\t\n",
            "\n",
            "Lorenzo Franceschi-Bicchierai \n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\t\tHackers have stolen email addresses, direct messages, and other personal data from users of two dating websites, according to a data breach expert. Earlier this week, someone alerted Troy Hunt, the fo\t\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\t\t\t\tMicrosoft makes its AI-powered Designer tool available in preview\t\t\t\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\t\tApr 27, 2023\t\n",
            "\n",
            "Kyle Wiggers \n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\t\tToday, Microsoft Designer, Microsoft’s AI-powered design tool, launched in public preview with an expanded set of features. Announced in October, Designer is a Canva-like web app that can genera\t\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\t\t\t\tBluesky’s best shot at success is to embrace shitposting\t\t\t\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\t\tApr 27, 2023\t\n",
            "\n",
            "Amanda Silberling \n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\t\tI have joined so many social media websites since October 27, 2022, the fateful day in which Elon Musk tanked his net worth and volunteered for the worst job in the world. There’s Mastodon, T2, \t\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\t\t\t\tPinterest announces multiyear ads partnership with Amazon alongside earnings beat\t\t\t\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\t\tApr 27, 2023\t\n",
            "\n",
            "Sarah Perez \n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\t\tPinterest today announced a multiyear strategic ad partnership with Amazon aimed at bringing more brands and relevant products to its platform. The new deal will make the e-commerce giant Pinterest&#8\t\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\t\t\t\tBrett Harrison ‘never lost faith in the business’ while at now-shuttered FTX\t\t\t\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\t\tApr 27, 2023\t\n",
            "\n",
            "Jacquelyn Melinek \n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\t\tAt Consensus 2023, Anthony Scaramucci and Brett Harrison shared their experiences during FTX’s downfall and what life has been like since.\t\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\t\t\t\tChief, a professional \u001b[0m\n",
            "Thought:\u001b[32;1m\u001b[1;3mHere is the response to your last comment:\n",
            "\n",
            "```json\n",
            "{\n",
            "    \"action\": \"Final Answer\",\n",
            "    \"action_input\": \"There is an article on TechCrunch titled 'Clubhouse needs to fix things, and today it cut more than half of staff' published on April 27, 2023.\"\n",
            "}\n",
            "```\n",
            "\n",
            "This information was obtained by searching the TechCrunch homepage for articles about Clubhouse.\u001b[0m\n",
            "\n",
            "\u001b[1m> Finished chain.\u001b[0m\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "\"There is an article on TechCrunch titled 'Clubhouse needs to fix things, and today it cut more than half of staff' published on April 27, 2023.\""
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            }
          },
          "metadata": {},
          "execution_count": 79
        }
      ],
      "source": [
        "conversational_agent.run(\"Is there an article about Clubhouse on https://techcrunch.com/? today\")"
      ]
    },
    {
      "cell_type": "code",
      "source": [
        "conversational_agent.run(\"What are the titles of the top stories on www.cbsnews.com/?\")"
      ],
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "4SSjIjI30rtZ",
        "outputId": "9373fedc-8a10-4d9c-f812-5aa635f657dd"
      },
      "id": "4SSjIjI30rtZ",
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "name": "stdout",
          "text": [
            "\n",
            "\n",
            "\u001b[1m> Entering new AgentExecutor chain...\u001b[0m\n",
            "\u001b[32;1m\u001b[1;3m```json\n",
            "{\n",
            "    \"action\": \"Get Webpage\",\n",
            "    \"action_input\": \"https://www.cbsnews.com/\"\n",
            "}\n",
            "```\u001b[0m\n",
            "Observation: \u001b[36;1m\u001b[1;3m\n",
            "CBS News - Breaking news, 24/7 live streaming news & top stories\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "Montana Rep. Zooey Zephyr\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "Disney vs. DeSantis\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "Chernobyl Anniversary\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "Freddie Mercury Auction\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "CBS News Live\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "Managing Your Money\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "Essentials Shopping\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "Newsletters\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "News\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "US\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "World\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "Politics\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "Entertainment\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "HealthWatch\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "MoneyWatch\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "CBS Village\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "Technology\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "Science\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "Crime\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "Sports\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "Essentials\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "Pence appears for 7 hours before grand jury\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "Proliferation of modified weapons cause for alarm, officials say\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "Two Army helicopters collide and crash in Alaska, killing three soldiers\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "E. Jean Carroll said she kept quiet about rape accusation out of \"fear\" of Trump\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "Democrats see abortion as winning issue with 2024 Biden reelection launch\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "Brittney Griner gives first press conference since Russian detainment\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "Trans, nonbinary state lawmakers press Biden on students athletics\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "Teens accused in Colorado rock-throwing death took \"memento\" photo of car\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "Lamar Jackson signs $260 million deal with Baltimore Ravens\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "Shows\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "Live\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "Local\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "More\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "Latest\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "Video\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "Photos\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "Podcasts\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "In Depth\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "Local\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "Global Thought Leaders\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "Innovators & Disruptors\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "Log In\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "Newsletters\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "Mobile\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "RSS\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "CBS Store\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "Paramount+\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "Join Our Talent Community\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "Davos 2023\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "Search\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "Search\n",
            "\n",
            "\n",
            "\n",
            "Login\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "Watch CBS News\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "Latest News\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "                          Pence appears for 7 hours before grand jury\n",
            "                    \n",
            "\n",
            "          Special counsel Jack Smith subpoenaed the former vice president in February.\n",
            "        \n",
            "3H ago\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "                          Proliferation of modified weapons cause for alarm, officials say\n",
            "                    \n",
            "\n",
            "          A simple conversion device can change a handgun to fire 15 rounds in under two seconds.\n",
            "        \n",
            "8H ago\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "Supreme Court cancels arguments in case over Title 42 border policy\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "                          Two Army helicopters collide and crash in Alaska, killing three soldiers\n",
            "                    \n",
            "\n",
            "          The choppers were returning from a training flight, the Army says. A fourth troop injured in the crash was hospitalized.\n",
            "        \n",
            "3H ago\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            " CBS News Live \n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "CBS News Live\n",
            "Live\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "                          E. Jean Carroll said she kept quiet about rape accusation out of \"fear\" of Trump\n",
            "                    \n",
            "\n",
            "          E. Jean Carroll has accused the former president of rape and then of defamation.\n",
            "        \n",
            "8H ago\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "                          Democrats see abortion as winning issue with 2024 Biden reelection launch\n",
            "                    \n",
            "\n",
            "          President Biden and Vice President Harris have signaled they'll put abortion access at the forefront of their reelection bid.\n",
            "        \n",
            "7H ago\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "60 Minutes visit to Carolyn Bryant Donham's home in 2004\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "                          Brittney Griner gives first press conference since Russian detainment\n",
            "                    \n",
            "\n",
            "          \"You're going to be faced with adversities in life,\" Griner said. \"This was a pretty big one. I just relied on my hard work to get through it.\"\n",
            "        \n",
            "10H ago\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "                          Trans, nonbinary state lawmakers press Biden on students athletics\n",
            "                    \n",
            "\n",
            "          The state lawmakers, including Montana Rep. Zooey Zephyr, want trans students to be able to \"fully participate in the sports of their choosing.\"\n",
            "        \n",
            "5H ago\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "                          Teens accused in Colorado rock-throwing death took \"memento\" photo of car\n",
            "                    \n",
            "\n",
            "          One of the suspects allegedly told the other two, \"We have to go back and see that\" after allegedly throwing the rock that killed Alexa Bartell, CBS Colorado reported.\n",
            "        \n",
            "6H ago\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "\n",
            "                          Lamar Jackson signs $260 million deal with Baltimore Ravens\n",
            "                    \n",
            "\n",
            "          The contract averages out to about $52 million per season, making Jackson the \u001b[0m\n",
            "Thought:\u001b[32;1m\u001b[1;3m```json\n",
            "{\n",
            "    \"action\": \"Final Answer\",\n",
            "    \"action_input\": \"Here are the titles of the top stories on CBS News:\\n\\n- Pence appears for 7 hours before grand jury\\n- Proliferation of modified weapons cause for alarm, officials say\\n- Two Army helicopters collide and crash in Alaska, killing three soldiers\\n- E. Jean Carroll said she kept quiet about rape accusation out of \\\"fear\\\" of Trump\\n- Democrats see abortion as winning issue with 2024 Biden reelection launch\\n- Brittney Griner gives first press conference since Russian detainment\\n- Trans, nonbinary state lawmakers press Biden on students athletics\\n- Teens accused in Colorado rock-throwing death took \\\"memento\\\" photo of car\\n- Lamar Jackson signs $260 million deal with Baltimore Ravens\"\n",
            "}\n",
            "```\u001b[0m\n",
            "\n",
            "\u001b[1m> Finished chain.\u001b[0m\n"
          ]
        },
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "'Here are the titles of the top stories on CBS News:\\n\\n- Pence appears for 7 hours before grand jury\\n- Proliferation of modified weapons cause for alarm, officials say\\n- Two Army helicopters collide and crash in Alaska, killing three soldiers\\n- E. Jean Carroll said she kept quiet about rape accusation out of \"fear\" of Trump\\n- Democrats see abortion as winning issue with 2024 Biden reelection launch\\n- Brittney Griner gives first press conference since Russian detainment\\n- Trans, nonbinary state lawmakers press Biden on students athletics\\n- Teens accused in Colorado rock-throwing death took \"memento\" photo of car\\n- Lamar Jackson signs $260 million deal with Baltimore Ravens'"
            ],
            "application/vnd.google.colaboratory.intrinsic+json": {
              "type": "string"
            }
          },
          "metadata": {},
          "execution_count": 78
        }
      ]
    },
    {
      "cell_type": "code",
      "source": [],
      "metadata": {
        "id": "hej1rBdW2U72"
      },
      "id": "hej1rBdW2U72",
      "execution_count": null,
      "outputs": []
    }
  ],
  "metadata": {
    "colab": {
      "provenance": []
    },
    "kernelspec": {
      "display_name": "langchain",
      "language": "python",
      "name": "langchain"
    },
    "language_info": {
      "codemirror_mode": {
        "name": "ipython",
        "version": 3
      },
      "file_extension": ".py",
      "mimetype": "text/x-python",
      "name": "python",
      "nbconvert_exporter": "python",
      "pygments_lexer": "ipython3",
      "version": "3.8.16"
    },
    "vscode": {
      "interpreter": {
        "hash": "578e1e8dce4dc6c542f1ea2d66a2d9db6ef592936dcc314004bdae386f827d38"
      }
    }
  },
  "nbformat": 4,
  "nbformat_minor": 5
}