{
  "cells": [
    {
      "cell_type": "markdown",
      "id": "8f210ec3",
      "metadata": {},
      "source": [
        "# Bash\n",
        "It can often be useful to have an LLM generate bash commands, and then run them. A common use case for this is letting the LLM interact with your local file system. We provide an easy util to execute bash commands."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 1,
      "id": "f7b3767b",
      "metadata": {},
      "outputs": [],
      "source": [
        "from langchain.utilities import BashProcess"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 2,
      "id": "cf1c92f0",
      "metadata": {},
      "outputs": [],
      "source": [
        "bash = BashProcess()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 3,
      "id": "2fa952fc",
      "metadata": {},
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "apify.ipynb\n",
            "arxiv.ipynb\n",
            "bash.ipynb\n",
            "bing_search.ipynb\n",
            "chatgpt_plugins.ipynb\n",
            "ddg.ipynb\n",
            "google_places.ipynb\n",
            "google_search.ipynb\n",
            "google_serper.ipynb\n",
            "gradio_tools.ipynb\n",
            "human_tools.ipynb\n",
            "ifttt.ipynb\n",
            "openweathermap.ipynb\n",
            "python.ipynb\n",
            "requests.ipynb\n",
            "search_tools.ipynb\n",
            "searx_search.ipynb\n",
            "serpapi.ipynb\n",
            "wikipedia.ipynb\n",
            "wolfram_alpha.ipynb\n",
            "zapier.ipynb\n",
            "\n"
          ]
        }
      ],
      "source": [
        "print(bash.run(\"ls\"))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 4,
      "id": "e7896f8e",
      "metadata": {},
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "apify.ipynb\n",
            "arxiv.ipynb\n",
            "bash.ipynb\n",
            "bing_search.ipynb\n",
            "chatgpt_plugins.ipynb\n",
            "ddg.ipynb\n",
            "google_places.ipynb\n",
            "google_search.ipynb\n",
            "google_serper.ipynb\n",
            "gradio_tools.ipynb\n",
            "human_tools.ipynb\n",
            "ifttt.ipynb\n",
            "openweathermap.ipynb\n",
            "python.ipynb\n",
            "requests.ipynb\n",
            "search_tools.ipynb\n",
            "searx_search.ipynb\n",
            "serpapi.ipynb\n",
            "wikipedia.ipynb\n",
            "wolfram_alpha.ipynb\n",
            "zapier.ipynb\n",
            "\n"
          ]
        }
      ],
      "source": [
        "bash.run(\"cd ..\")\n",
        "# The commands are executed in a new subprocess each time, meaning that\n",
        "# this call will return the same results as the last.\n",
        "print(bash.run(\"ls\"))"
      ]
    },
    {
      "attachments": {},
      "cell_type": "markdown",
      "id": "851fee9f",
      "metadata": {},
      "source": [
        "## Terminal Persistance\n",
        "\n",
        "By default, the bash command will be executed in a new subprocess each time. To retain a persistent bash session, we can use the `persistent=True` arg."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 5,
      "id": "4a93ea2c",
      "metadata": {},
      "outputs": [],
      "source": [
        "bash = BashProcess(persistent=True)"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 6,
      "id": "a1e98b78",
      "metadata": {},
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "custom_tools.ipynb\t\tmulti_input_tool.ipynb\n",
            "examples\t\t\ttool_input_validation.ipynb\n",
            "getting_started.md\n"
          ]
        }
      ],
      "source": [
        "bash.run(\"cd ..\")\n",
        "# Note the list of files is different\n",
        "print(bash.run(\"ls\"))"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": null,
      "id": "e13c1c9c",
      "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": "3.8.16"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 5
}