Init.ai CLI

A suite of tools for developing Projects on the init.ai platform.

Note: This CLI is currently an alpha release and subject to change at any time.

Pull Requests and Contributions

At the current time, this project is not developed openly on Github. We hope to change that soon, but in the meantime, any PRs will have to be reviewed and manually applied by Init.ai team members.

We will provide a list of contributors and make sure to acknowledge all feedback we receive.

Note: Currently, this project is not distributed under a permissive open source license. Contributions will be accepted with the understanding that they become part of our IP. This is likely to change in the future and we will update this repository accordingly at that time.

Installation

Make sure you have Node.js installed. We recommend using version 4.3.2 (see below). If you are using a tool such as nvm simply run:

$ nvm install 4.3.2
$ nvm use 4.3.2

$ npm i -g initai-cli

Installing this CLI globally will make the iai namespeace available to you from the command line. If you, however would prefer to not install this package globally, you may install it in your current project and run it as follows:

# On Mac

node ./node_modules/initai-cli/programs/iai-watch

# On Windows

node .\node_modules\initai-cli\programs\iai-watch

Note: It is recommended that you use Node.js version 4.3.2 to ensure you are running the same version of Node that is available in AWS Lambda. While the developer tools do not currently interact with or represent Lambda, upcoming features will require this version.

Uninstall

If installed globally:

$ npm uninstall -g init-cli

Usage

Conversation Training Server

The conversation training server acts as a bridge between your production console and your local file system. This will start a server on your local machine that can communicate with the console running in your browser. It does this by facilitating a websocket connection between the two allowing changes to conversation data in the browser to be persisted locally via CML files.

To run the server, make sure you are in the directory of your working project (cd /path/to/my-project) and run the server command:

$ iai watch

You will notice the console will reflect successful connection status immediately. You can of course, at any time, kill this server and the browser will no longer have access to any locally running codebase.

Note: The current version of the CLI does not automatically switch directories when you switch projects in the console. When switching between projects in the console, you must hop back to your terminal, cd into the new project directory and restart the CLI.

Help

$ iai --help
$ iai -h

List available commands

$ iai

Supported Platforms

• Mac OS X 10.10+
• Windows 8.1+
• Linux