sem-cli

• This is a terminal app which works with human language commands.
• It is developed in Node.js without any framework or external library.
• This is the client side of the project. There is not so much logic in it. You can find the server side of it in there: https://github.com/hcoz/sem-cli-serverless

Installation

You can install as a npm package. I recommend global install but it is up to you!

npm install sem-cli -g

Commands

• sem-look: It is for showing the corresponding command for your message.

sem-look <your message in human language>

You can use the all languages which are supported by Wit.ai To check list of all languages, visit https://wit.ai/faq

Here is the example:

sem-look list files

You can also add arguments to your messages by using this notation p="<argument>"

Here is the example:

sem-look compare files p="a.txt b.txt"

It will return Your command: "cmp a.txt b.txt" with danger level: "low" for MacOS

• sem-exec: It is for executing the corresponding command for your message.

sem-exec <your message in human language>

You can use the all languages which are supported by Wit.ai To check list of all languages, visit https://wit.ai/faq

Here is the example:

sem-exec list files

You can also add arguments to your messages by using this notation p="<argument>"

Here is the example:

sem-exec list files p="subdir"

It will show the files under <your current directory>/<subdir>

• sem-suggest: It is for suggesting a new (intent, command, dangerLevel) triple to extend the database
  • intent: It is a short explanation for the purpose of this command. It will be the key of this triple
  • command: It is the related command. (Please suggest a command which is working on your current operating system)
  • dangerLevel: It shows how dangerous to run this command. It can be "low", "medium" or "high". ("high" dangerous commands will not be run before a client approval)

The order of parameters is important. Please use same as the shown below.

sem-suggest intent="<intent>" command="<command>" dangerLevel="<dangerLevel>"

Here is the example:

sem-suggest intent="list_files" command="ls" dangerLevel="low"