GPT3 Powered CLI

Installation

You'd need to install it globally

npm i @abhagsain/ai-cli -g

Usage

$ ai ask "Check process running on port"

You'd need to enter your own OpenAI API key Here's how you can get one

1. Go to https://openai.com/api/login
2. Create an account or log into your existing account
3. Go to https://beta.openai.com/account/api-keys or
4. Run ai auth, enter your API KEY and you're good to go!

Pricing

The current prompt length is ~800 tokens and average response length is ~40 tokens. The pricing for gpt-3.5-turbo is $0.001 per 1K input tokens and $0.002 per 1K output tokens which is ~$0.0009/command. We'll see if we can improve the response as well as reduce the per-command-cost with fine-tuning.

---

Add autocomplete by running ai autocomplete and follow the instructions. It's super easy.

__

Liked this project? Checkout my other side project. SlidesAI.io - AI Presentation Maker

__

Auto generated documentation

• Usage
• Commands

Usage

$ npm install -g @abhagsain/ai-cli
$ ai COMMANDrunning command...
$ ai (--version)@abhagsain/ai-cli/1.3.0 darwin-arm64 node-v18.0.0
$ ai --help [COMMAND]USAGE $ ai COMMAND...

Commands

• ai ask [question]
• ai auth
• ai autocomplete [SHELL]
• ai help [COMMAND]
• ai model
• ai update [CHANNEL]

ai ask [question]

Ask question to GPT3 from your terminal

USAGE
$ ai ask [question]
ARGUMENTS
QUESTION Your question
DESCRIPTION
Ask question to GPT3 from your terminal
EXAMPLES
$ ai ask "Check running process on port 3000"

See code: dist/commands/ask.ts

ai auth

Update existing or add new OpenAI API Key

USAGE
$ ai auth
DESCRIPTION
Update existing or add new OpenAI API Key
EXAMPLES
$ ai auth (Follow the prompt)

See code: dist/commands/auth.ts

ai autocomplete [SHELL]

display autocomplete installation instructions

USAGE
$ ai autocomplete [SHELL] [-r]
ARGUMENTS
SHELL shell type
FLAGS
-r, --refresh-cache Refresh cache (ignores displaying instructions)
DESCRIPTION
display autocomplete installation instructions
EXAMPLES
$ ai autocomplete
$ ai autocomplete bash
$ ai autocomplete zsh
$ ai autocomplete --refresh-cache

See code: @oclif/plugin-autocomplete

ai help [COMMAND]

Display help for ai.

USAGE
$ ai help [COMMAND] [-n]
ARGUMENTS
COMMAND Command to show help for.
FLAGS
-n, --nested-commands Include all nested commands in the output.
DESCRIPTION
Display help for ai.

See code: @oclif/plugin-help

ai model

Change model preference (default: gpt-3.5-turbo)

USAGE
$ ai model
DESCRIPTION
Change model preference (default: gpt-3.5-turbo)
EXAMPLES
$ ai model (Follow the prompt)

See code: dist/commands/model.ts

ai update [CHANNEL]

update the ai CLI

USAGE
$ ai update [CHANNEL] [-a] [-v <value> | -i] [--force]
FLAGS
-a, --available Install a specific version.
-i, --interactive Interactively select version to install. This is ignored if a channel is provided.
-v, --version=<value> Install a specific version.
--force Force a re-download of the requested version.
DESCRIPTION
update the ai CLI
EXAMPLES
Update to the stable channel:
$ ai update stable
Update to a specific version:
$ ai update --version 1.0.0
Interactively select version:
$ ai update --interactive
See available versions:
$ ai update --available

See code: @oclif/plugin-update