Terminal AI

Terminal AI is a command-line interface (CLI) tool that allows you to utilize the AI models provided by OpenAI in an easy to use way. It allows you to ask any kind of questions and generate content (text, images and audio) straight from the terminal. This tool defaults to using GPT-4, one of OpenAI's models.

Prerequisites

• Node.js installed in the system
• OpenAI API keys

Setup

Start by copying config.example.js to config.js

cp config.example.js config.js

Next you need to add your OpenAI API key to config.js.

Then install dependencies:

npm install

Installing Globally

Run npm install -g from the terminal-ai root directory. This makes the terminal-ai command available anywhere.

TIP: I like to set an alias in my .zshrc like so: alias ai="terminal-ai" so I don't have to type out "terminal-ai" every time.

Usage

You can run terminal-ai without any additional arguments to get a list of available sub commands.

ask

The ask subcommand allows you to ask terminal-ai a question and recieve an answer.

terminal-ai ask "What is the meaning of life?" -o output.txt

This will ask the question "What is the meaning of life?" to the AI model and output the response to the output.txt file.

terminal-ai ask "What is the meaning of life?" -f input.txt

This will read the input.txt file and ask a question based on it.

imagine

The imagine subcommand can be used to generate images from a text prompt.

speak

The speak subcommand allows you to generate audio from text

Commands

1. ask: Ask a question. This command takes a question as an argument and optionally an output file and an input file.

• -o, --output <output>: Output the response to a file
• -f, --file <file>: An input file to ask the question about

terminal-ai ask "What is the meaning of life?" -o output.txt -f input.txt

---

I hope you find this tool beneficial to your day to day work.