ai.py

A single-file Python script that interacts with ChatGPT API in the command-line.

Features:

• Use shortcuts to access predefined prompts
• Highlight code in output
• Support one-shot queries and conversations
• Use special command like !set to control the behavior when chatting

Install

Just copy the script to a folder in $PATH, like /usr/local/bin. You can also change its name to ai to get ride of the .py extension.

Here's a command that can directly install the script into your system:

curl https://raw.githubusercontent.com/reorx/ai.py/master/ai.py -o /usr/local/bin/ai && chmod +x /usr/local/bin/ai

You can also install it with pip or pipx:

pip install aidotpy

Usage

Paste your OpenAI API key to ~/.config/ai-py/config.json, or set it in AI_PY_API_KEY environment variable.

mkdir -p ~/.config/ai-py
echo '{"api_key":"<Your API key>"}' > ~/.config/ai-py/config.json

For detail usage of the script, please read the description of ./ai.py -h:

usage: ai [-h] [-s SYSTEM] [-c] [--history HISTORY] [-w] [-v] [-t] [-d] [--version] [PROMPT]

A simple CLI for ChatGPT API

positional arguments:
  PROMPT                your prompt, leave it empty to run REPL. you can use @ to load prompt
                        from the prompts file.

options:
  -h, --help            show this help message and exit
  -s SYSTEM, --system SYSTEM
                        system message to use at the beginning of the conversation. if starts
                        with @, the message will be located through the prompts file
  -c, --conversation    enable conversation, which means all the messages will be sent to the
                        API, not just the last one. This is only useful to REPL
  --history HISTORY     load the history from a JSON file.
  -w, --write-history   write new messages to --history file after each chat.
  -v, --verbose         verbose mode, show execution info and role in the message
  -t, --show-tokens     show a breakdown of the tokens used in the prompt and in the response
  -d, --debug           debug mode, enable logging
  --version             show program's version number and exit

One-off query

Pass the prompt as the first argument:

./ai.py 'hello world'

You can also pass the prompt through a pipe (|):

head README.md | ./ai.py 'Proofreading the following text:'

REPL

Run without argument for Read–eval–print loop:

./ai.py

By default only the last message and the system message are sent to the API, if you want it to remember all the context (i.e. send all the messages in each chat), add -c argument to enable conversation:

./ai.py -c

System message

You can pass a system message to define the behavior for the assistant:

./ai.py -s 'You are a proofreader' 'its nice know you'

You can also save your predefined system messages in ~/.config/ai-py/prompts.json and refer them with @ at the beginning, this will be covered in the next section.

Prompt shortcuts

You can predefine prompts in ~/.config/ai-py/prompts.json and refer to them by using @ as a prefix. This works for both system messages and user messages.

Suppose your ~/.config/ai-py/prompts.json looks like this:

{
  "system": {
    "cli": "As a technology assistant with expertise in command line, answer questions in simple and short words for users who have a high-level background. Provide only one example, and explain as less as possible."
  },
  "user": {
    "native": "Paraphrase the following sentences to make it more native:\n",
    "revise": "Revise the following sentences to make them more clear concise and coherent:\n",
    "": ""
  }
}

Then you can use the cli prompt shortcut in system message by:

./ai.py -s @cli

and use the native or revise prompt shortcut in user message by:

./ai.py '@native its nice know you'

It's great to get to know you.

Verbose mode

Add -v to print role name and parameters used in the API call.

Special commands

You can use special commands to control the behavior of the script when running in REPL.

Here's a list of available commands:

• !set <key> <value>: set a key-value pair in the config, available keys are:
  • verbose: set to True or False, e.g. !set verbose True
  • conversation: set to True or False, e.g. !set conversation True
  • system: set the system message. e.g. !set system you are a poet, !set system @cli
  • params: set the parmeters for the ChatGPT API. e.g. !set params temperature 0.5
  • model: set the model to use. e.g. !set model gpt-4
• !info: print the execution info
• !write-history: write current messages to history file. e.g. !write-history history.json