Natsuko supports writing scripts to power its automatic moderation engine. Scripts are written in JavaScript. This allows you to, quite literally, teach the bot your server's rules. They are executed on message. Every object is wrapped in a Safe* object, which can all be found in the ScriptEngine package.

Command Usage

Now, before we get ahead of ourselves lets go over how the n;script command is operated. There are 4 subcommands, show, add, edit, delete . n;script show [id]: Checks the amount of script slots you have available (as of yet, the limit is 3!) Or, returns the contents of a script by its id. n;script add <script>: Adds a new script, if there is space. n;script edit <id> <script>: Edits an existing script, by id. n;script delete <id>: Deletes an existing script, by id.

Examples

Naughty Word Filter

By the end of this example, we would have learnt how to delete messages, how to log messages, and finally how to punish users. This is a pretty simple script, so this should be easy for you to adapt yourself.

Let's say one of our rules is that you cannot say the phrase "naughty word". We can easily write a script for this like so:

naughty-word-filter.js
if (message.getContents().contains("naughty word")) {
    message.delete();
}

Alright, lets break this script down. message is the message that was sent. It is implemented in Java as a SafeMessage object, which is a wrapper around a standard message. It has the following methods:

message.getContents(); // returns a string containing the contents of the message
message.getTimestamp(); // returns a Java Instant object containing the message timestamp
message.getId(); // returns a Snowflake containing the ID
message.getAuthor(); // returns a SafeMember object which is the member that sent the message
message.getChannel(); // returns a SafeChannel object which is the channel the message was sent in
message.getGuild(); // returns a SafeGuild object which is the guild the message was sent in
message.mentionsEveryone(); // returns if the message mentions everyone
message.delete(); // deletes the message
message.edit('new content'); // edits the message ONLY if it is sent by the bot and returns the SafeMessage object

Let's do another example:

logging-naughty-word-filter.js
if (message.getContents().contains("naughty word")) {
    message.getGuild().getChannelById(util.id("591735345406672935")).send(
        message.getAuthor().getName() + " said a naughty word!"
    );
}

This example does not delete the message, but instead logs the event in the channel with the ID of 591735345406672935. As documented earlier, getGuild() returns the guild for the message, which allows you to then make your own. The channel 591735345406672935could be an audit log channel. A new method has been introduced here, called id. This method creates a Snowflake object so that you can get channels and the like.

Maybe we want to add some punishments into the mix?

punishing-naughty-word-filter.js
if (message.getContents().contains("naughty word")) {
    message.getAuthor().ban("posted a naughty word!", 7);
}

This is a punishing version of our naughty word filter. This bans the author of the message for "posted a naughty word!" and then removes the last 7 days of messages.

Let's put everything we've learnt together:

ultimate-naughty-word-filter.js
if (message.getContents().contains("naughty word")) {
    message.getGuild().getChannelById(util.id("591735345406672935")).send(
        message.getAuthor().getName()
        + " sent a naughty word, so I deleted it and banned them!"
    );
    message.getAuthor().ban("said a naughty word!", 0); // let's be forgiving
    message.delete();
}

And here's the final rundown.

• Check the message to see if it has "naughty word" in it.

• Get the guild the message was posted in.

• Get a channel in the guild by ID "591735345406672935".

• Post a message saying "Username sent a naughty word, so I deleted it and banned them!".

• Get the message author and ban them with the reason "said a naughty word!", but we don't delete any messages from them so that their message history isn't lost.

• Delete the message.

Of course you can do much more advanced things, but that is out of the scope of this guide.