smartchat

Interact with ChatGPT from the command-line.

smartchat is an npm package that allows you to interact with ChatGPT, a powerful language model developed by OpenAI, directly from the command line. With this package, you can have conversational interactions with the model, generating responses to your queries and engaging in a chat-like experience.

Features

Chat with ChatGPT from the command-line
Support both the official ChatGPT API and the unofficial proxy API of ChatGPT API
Supports both Q&A and Chat modes
One authorization, no need to re-authorize within a short period of time
Node based implementation
Rich tips, colored symbols for various log levels
Conversation stored in a local file

Install

$ npm install smartchat -g

Note: Make sure you're using node >= 18 so fetch is available (or node >= 14 if you install a fetch polyfill).

Usage

Before use smartchat, you may need to pass the 'Authorization'.

Authorization

$ smartchat auth

By default, it will choose the official ChatGPT API and then you have to provide an OpenAI API key for authorization.

D:\git-project\smartChat>smartchat auth
? Please enter a OpenAI API key <your OpenAI API key>
√ Authorization is successful, please use command "smartchat <prompt>" to start a conversation with chatgpt

D:\git-project\smartChat>

Alternatively, you can use --no-offical to choose an unofficial proxy API and provide an Access Token for authorization.

D:\git-project\smartChat>smartchat auth --no-offical
? Please enter a OpenAI Access Token <your OpenAI Access Token>
√ Authorization is successful, please use command "smartchat <prompt>" to start a conversation with chatgpt

D:\git-project\smartChat>

Q&A

In Q&A mode, after you provide a prompt, chatgpt will give you the result and then immediately exit automatically.

$ smartchat <prompt>

D:\git-project\smartChat>smartchat "Who are you ?"
prompt: Who are you ?

 I am ChatGPT, a language model developed by OpenAI. I'm designed to assist with a wide range of questions and provide
 information on various topics. How can I assist you today?

D:\git-project\smartChat>

Chat

Chat mode allows you to talk to chatgpt continuously in real time, you can use exit to exit the conversation.

$ smartchat <prompt> --chat

D:\git-project\smartChat>smartchat "Who are you ?" --chat
++++++++++++++++++++++++++++++++++++++++++++++++++++++++
You are chatting with chatgpt, you can quit with "exit".
++++++++++++++++++++++++++++++++++++++++++++++++++++++++
prompt: Who are you ?

 I am ChatGPT, an AI language model developed by OpenAI. I'm designed to assist with answering questions, providing
 information, and engaging in conversations on a wide range of topics. How can I assist you today?


? next prompt:  "What can you assist me ?"

 I can assist you with a wide range of topics and tasks. Here are a few examples:

  1. Answering questions: I can provide information on various subjects, including science, history, geography, sports,
     and more. Just ask me anything you'd like to know!

  2. Writing assistance: If you need help with writing, whether it's for academic purposes, creative writing, or
     professional documents, I can offer suggestions, grammar and style corrections, and help you improve your overall
     writing quality.

    .....

 Please let me know what specific assistance you require, and I'll do my best to help you!

? next prompt:

Deauthorization

$ smartchat deauth

When you execute the above command, your OpenAI API key or Access Token will be removed from this local config file.

Cli

By default, the results are stored in a local config file, and every invocation starts a new conversation. You can use -c to continue the previous conversation.

Usage:
  $ smartchat <prompt>

Commands:
  auth      Set a openai api key or access token
  deauth    Remove openai api key or access token
  <prompt>  Ask ChatGPT a question
  rm-cache  Clear the local message cache
  ls-cache  Output the local message cache file path

For more info, run any command with the `--help` flag:
  $ smartchat auth --help
  $ smartchat deauth --help
  $ smartchat --help
  $ smartchat rm-cache --help
  $ smartchat ls-cache --help

Options:
  -c, --continue       Continue last conversation (default: false)
  -s, --store          Enables the local message cache (default: true)
  --chat               Continuous dialogue mode (default: false)
  -p, --proxy <proxy>  Provide a reverse proxy for unofficial proxy API (default: https://ai.fakeopen.com/api/conversation)
  -m, --model <model>  Model for the official ChatGPT API (default: gpt-3.5-turbo)
  -h, --help           Display this message
  -v, --version        Display version number

About the official ChatGPT API and the unofficial proxy API

Method	Free?	Robust?	Quality?
`ChatGPTAPI`	❌ No	✅ Yes	✅️ Real ChatGPT models + GPT-4
`ChatGPTUnofficialProxyAPI`	✅ Yes	❌ No️	✅ ChatGPT webapp

ChatGPTAPI - Uses the gpt-3.5-turbo model with the official OpenAI chat completions API (official, robust approach, but it's not free)
ChatGPTUnofficialProxyAPI - Uses an unofficial proxy server to access ChatGPT's backend API in a way that circumvents Cloudflare (uses the real ChatGPT and is pretty lightweight, but relies on a third-party server and is rate-limited)

Access Token for unofficial proxy API

Reverse Proxy

By default, the reverse proxy is https://ai.fakeopen.com/api/conversation for the unofficial proxy API, You can override the reverse proxy:

$ smartchat <prompt> --proxy <url>

Known reverse proxies run by community members include:

Reverse Proxy URL	Author	Rate Limits	Last Checked
`https://ai.fakeopen.com/api/conversation`	@pengzhile	5 req / 10 seconds by IP	4/18/2023
`https://api.pawan.krd/backend-api/conversation`	@PawanOsman	50 req / 15 seconds (~3 r/s)	3/23/2023

Note: info on how the reverse proxies work is not being published at this time in order to prevent OpenAI from disabling access.

Access Token

To use unofficial proxy API, you'll need an OpenAI access token from the ChatGPT webapp. To do this, you can use any of the following methods which take an email and password and return an access token:

Node.js libs

These libraries work with email + password accounts (e.g., they do not support accounts where you auth via Microsoft / Google).

Alternatively, you can manually get an accessToken by logging in to the ChatGPT webapp and then opening https://chat.openai.com/api/auth/session, which will return a JSON object containing your accessToken string:

{
    "user": {
        "id": " ",
        "name": " ",
        "email": " ",
        "image": " ",
        "picture": " ",
        "idp": " ",
        "iat":  ,
        "mfa": false,
        "groups": [],
        "intercom_hash": ""
    },
    "expires": "2023-06-27T14:22:43.040Z",
    "accessToken": "",
    "authProvider": ""
}

Access tokens last for days.

Contributing

Contributions are welcome! If you encounter any issues or have suggestions for improvements, please feel free to open an issue or submit a pull request on the GitHub repository. When contributing, please ensure to follow the existing coding style and guidelines.

Note

Because of the way Terminals are built, it is not possible to update a text outside the viewport of the terminal. So for the time being, it is not possible to achieve real-time typing output in the terminal.

We strongly recommend the official ChatGPT API since it use the officially supported API from OpenAI. The unofficial proxy API may be removed in a future release.

Using a reverse proxy will expose your access token to a third-party. There shouldn't be any adverse effects possible from this, but please consider the risks before using this method.

Author

GitHub @zhangqiangzgz

zhangqiangzgz / smartchat