The core foundation of a server on Discord is the community that populates it. Your community is what you engage with, protect, and grow over time. Engagement is important to focus on, but it’s just as important to make sure you are facilitating positive and welcoming engagement.
Positive engagement can mean a lot of things, but in this article, we will be referring to the way in which moderation can affect the culture of the server you are moderating. As moderators your policies, knowledge of your community, and deductive skills influence the way in which your community engages with each other and with your team.
When you establish and nurture your community, you are growing a collective group of people who all enjoy at least some of the same things. Regardless of your server topic, you are undoubtedly going to have members across different a variety of ethnicities, sexual orientations, and identities from across the world. Ensuring that your space on Discord is a space for them to belong necessitates making it safe for them to feel like they can be themselves, wholly, and without reservation. Your members are all humans, all community members, all people that deserve respect and deserve to be welcomed.
When you are establishing your community, it’s important to have a basic understanding of what kind of environment you would like your server to be. It’s good to break down the general moderation philosophy on what content and discussion you’d like your community to engage in and what content would be inappropriate given the space. Depending on the topic of your server these goals may be different, but some common questions you can ask to establish general boundaries are:
*Unless you are using the channel description for verification instructions rather than an automatic greeter message.
If you want to use the remove unverified role method, you will need a bot that can automatically assign a role to a user when they join.
Once you decide whether you want to add or remove a role, you need to decide how you want that action to take place. Generally, this is done by typing a bot command in a channel, typing a bot command in a DM, or clicking on a reaction. The differences between these methods are shown below.
In order to use the command in channel method, you will need to instruct your users to remove the Unverified role or to add the Verified role to themselves.
This section will be more specific and will break down the most common ways in which a user can engage in harmful rhetoric, how to de-escalate discussions that attack marginalized communities, and how to properly address uncommon symbols used to attack communities.
Harmful Terms and Ableist Language
Members of your community may use obscure symbols or terms to send an offensive message while avoiding any blatant attention or triggering filters. While some of these will be used under the guise of being “internet culture” it’s important to be cognizant that these symbols and language can cause a lot of pain to many marginalized people. Understanding and approaching these terms seriously will help mitigate any long-term server culture damage and personal damage this kind of behavior can cause. This will not only be symbols, but popular terms that are used to cause intentional discourse and push harmful narratives.
These are all terms and symbols used to specifically target and belittle groups of people, and are harmful to the growth of a welcoming environment for your community. Some are slurs, while some are generally harmful rhetoric;
LGBTQ+ Specific Slurs
While some of these terms may be popular in certain spaces (such as gaming), it’s important to understand the history and weight behind them, and think accordingly about their place in your server long-term.
Markdown is also supported in an embed. Here is an image to showcase an example of these properties:
Example image to showcase the elements of an embed
An important thing to note is that embeds also have their limitations, which are set by the API. Here are some of the most important ones you need to know:
An important thing to note is that embeds also have their limitations, which are set by the API. Here are some of the most important ones you need to know:
If you feel like experimenting even further you should take a look at the full list of limitations provided by Discord here.
It’s very important to keep in mind that when you are writing an embed, it should be in JSON format. Some bots even provide an embed visualizer within their dashboards. You can also use this embed visualizer tool which provides visualization for bot and webhook embeds.
Online platforms tend to have a large amount of hateful content and rhetoric used against marginalized groups of people. When crafting a community, there has to be a common goal of acceptance and welcoming that you provide for all of your members. In online communities, it is not uncommon for users to voice their disdain at other users for their choices in pronouns, gender presentation, or anything that relates them to the LGBTQ+ community.
An ally is someone who is not a part of the LGBTQ+ umbrella who supports and ‘allies’ with the community to create an open and welcoming atmosphere in your server. It is important to understand your role in being an ally to your community and to your users.
Pronouns are what people use to refer to a person without directly stating their name. Common pronouns are They/Them, She/Her, and He/Him. There are many others that are not covered here, but pronouns can be very important to someone’s identity and how they’d prefer to be addressed. Users, at one point or another, may make jokes such as calling users incorrect pronouns intentionally to invalidate their identity, dehumanize them, and humiliate members of the community out of their ability to interact within the server.
Additionally, it’s worth noting that comments such as ‘there are only two genders’ are used to directly disrespect and undermine the trans community, without directly appearing confrontational. This is used to skirt by rules by appearing to be much less antagonizing than these words truly are.
There are a few LGBTQ+ specific terms that are good to be aware of when interacting with LGBTQ+ members of your community. It’s important to also keep up with your communities- it can be helpful to do some research as terms and issues pop up. Quickly Googling a new term that you come across from a member can make them feel much more welcome in the community as a whole and address sore spots as well.
These are also some terms to be aware of that are related to LGBTQ+ issues that you may see being brought up:
Even though this comparison is important for better understanding of both bots and webhooks, it does not mean you should limit yourself to only picking one or the other. Sometimes, bots and webhooks work their best when working together. It’s not uncommon for bots to use webhooks for logging purposes or to distinguish notable messages with a custom avatar and name for that message. Both tools are essential for a server to function properly and make for a powerful combination.
*Unconfigurable filters, these will catch all instances of the trigger, regardless of whether they’re spammed or a single instance
**Gaius also offers an additional NSFW filter as well as standard image spam filtering
***YAGPDB offers link verification via google, anything flagged as unsafe can be removed
****Giselle combines Fast Messages and Repeated Text into one filter
Anti-Spam is integral to running a large private server, or a public server. Spam, by definition, is irrelevant or unsolicited messages. This covers a wide base of things on Discord, there are multiple types of spam a user can engage in. The common forms are listed in the table above. The most common forms of spam are also very typical of raids, those being Fast Messages and Repeated Text. The nature of spam can vary greatly but the vast majority of instances involve a user or users sending lots of messages with the same contents with the intent of disrupting your server.
There are subsets of this spam that many anti-spam filters will be able to catch. If any of the following: Mentions, Links, Invites, Emoji, and Newline Text are spammed repeatedly in one message or spammed repeatedly across several messages, they will provoke most Repeated Text and Fast Messages filters appropriately. Subset filters are still a good thing for your anti-spam filter to contain as you may wish to punish more or less harshly depending on the spam. Namely, Emoji and Links may warrant separate punishments. Spamming 10 links in a single message is inherently worse than having 10 emoji in a message.
Anti-spam will only act on these things contextually, usually in an X in Y fashion where if a user sends, for example, 10 links in 5 seconds, they will be punished to some degree. This could be 10 links in one message, or 1 link in 10 messages. In this respect, some anti-spam filters can act simultaneously as Fast Messages and Repeated Text filters.
Sometimes, spam may happen too quickly for a bot to catch up. There are rate limits in place to stop bots from harming servers that can prevent deletion of individual messages if those messages are being sent too quickly. This can often happen in raids. As such, Fast Messages filters should prevent offenders from sending messages; this can be done via a mute, kick or ban. If you want to protect your server from raids, please read on to the Anti-Raid section of this article.
Text filters allow you to control the types of words and/or links that people are allowed to put in your server. Different bots will provide various ways to filter these things, keeping your chat nice and clean.
*Defaults to banning ALL links
**YAGPDB offers link verification via google, anything flagged as unsafe can be removed
***Setting a catch-all filter with carl will prevent link-specific spam detection
A text filter is integral to a well moderated server. It’s strongly, strongly recommended you use a bot that can filter text based on a blacklist. A Banned words filter can catch links and invites provided http:// and https:// are added to the word blacklist (for all links) or specific full site URLs to block individual websites. In addition, discord.gg can be added to a blacklist to block ALL Discord invites.
A Banned Words filter is integral to running a public server, especially if it’s a Partnered, Community or Verified server, as this level of auto moderation is highly recommended for the server to adhere to the additional guidelines attached to it. Before configuring a filter, it’s a good idea to work out what is and isn’t ok to say in your server, regardless of context. For example, racial slurs are generally unacceptable in almost all servers, regardless of context. Banned word filters often won’t account for context, with an explicit blacklist. For this reason, it’s also important a robust filter also contains whitelisting options. For example, if you add the slur ‘nig’ to your filter and someone mentions the country ‘Nigeria’ they could get in trouble for using an otherwise acceptable word.
Filter immunity may also be important to your server, as there may be individuals who need to discuss the use of banned words, namely members of a moderation team. There may also be channels that allow the usage of otherwise banned words. For example, a serious channel dedicated to discussion of real world issues may require discussions about slurs or other demeaning language, in this exception channel based Immunity is integral to allowing those conversations.
Link filtering is important to servers where sharing links in ‘general’ chats isn’t allowed, or where there are specific channels for sharing such things. This can allow a server to remove links with an appropriate reprimand without treating a transgression with the same severity as they would a user sending a racial slur.
Whitelisting/Blacklisting and templates for links are also a good idea to have. While many servers will use catch-all filters to make sure links stay in specific channels, some links will always be malicious. As such, being able to filter specific links is a good feature, with preset filters (Like the google filter provided by YAGPDB) coming in very handy for protecting your user base without intricate setup however, it is recommended you do configure a custom filter to ensure specific slurs, words etc. that break the rules of your server, aren’t being said.
Invite filtering is equally important in large or public servers where users will attempt to raid, scam or otherwise assault your server with links with the intention of manipulating your user base to join or where unsolicited self-promotion is potentially fruitful. Filtering allows these invites to be recognized, and dealt with more harshly. Some bots may also allow by-server white/blacklisting allowing you to control which servers are ok to share invites to, and which aren’t. A good example of invite filtering usage would be something like a partners channel, where invites to other, closely linked, servers are shared. These servers should be added to an invite whitelist to prevent their deletion.
Raids, as defined earlier in this article, are mass-joins of users (often selfbots) with the intent of damaging your server. There are a few methods available to you in order for you to protect your community from this behavior. One method involves gating your server with verification appropriately, as discussed in DMA 301.You can also supplement or supplant the need for verification by using a bot that can detect and/or prevent damage from raids.
*Unconfigurable, triggers raid prevention based on user joins & damage prevention based on humanly impossible user activity. Will not automatically trigger on the free version of the bot.
Raid detection means a bot can detect the large number of users joining that’s typical of a raid, usually in an X in Y format. This feature is usually chained with Raid Prevention or Damage Prevention to prevent the detected raid from being effective, wherein raiding users will typically spam channels with unsavoury messages.
Raid-user detection is a system designed to detect users who are likely to be participating in a raid independently of the quantity of frequency of new user joins. These systems typically look for users that were created recently or have no profile picture, among other triggers depending on how elaborate the system is.
Raid prevention stops a raid from happening, either by Raid detection or Raid-user detection. These countermeasures stop participants of a raid specifically from harming your server by preventing raiding users from accessing your server in the first place, such as through kicks, bans, or mutes of the users that triggered the detection.
Damage prevention stops raiding users from causing any disruption via spam to your server by closing off certain aspects of it either from all new users, or from everyone. These functions usually prevent messages from being sent or read in public channels that new users will have access to. This differs from Raid Prevention as it doesn’t specifically target or remove new users on the server.
Raid anti-spam is an anti spam system robust enough to prevent raiding users’ messages from disrupting channels via the typical spam found in a raid. For an anti-spam system to fit this dynamic, it should be able to prevent Fast Messages and Repeated Text. This is a subset of Damage Prevention.
Raid cleanup commands are typically mass-message removal commands to clean up channels affected by spam as part of a raid, often aliased to ‘Purge’ or ‘Prune’.It should be noted that Discord features built-in raid and user bot detection, which is rather effective at preventing raids as or before they happen. If you are logging member joins and leaves, you can infer that Discord has taken action against shady accounts if the time difference between the join and the leave times is extremely small (such as between 0-5 seconds). However, you shouldn’t rely solely on these systems if you run a large or public server.
Messages aren’t the only way potential evildoers can present unsavoury content to your server. They can also manipulate their Discord username or Nickname to cause trouble. There are a few different ways a username can be abusive and different bots offer different filters to prevent this.
*Gaius can apply same blacklist/whitelist to names as messages or only filter based on items in the blacklist tagged %name
**YAGPDB can use configured word-list filters OR a regex filter
Username filtering is less important than other forms of auto moderation, when choosing which bot(s) to use for your auto moderation needs, this should typically be considered last, since users with unsavory usernames can just be nicknamed in order to hide their actual username.
When it comes to the content you allow or moderate in your server, it’s important to, again, reflect on what type of community you are. It’s also important that you act quickly and precisely on this type of harmful behavior. Some users will slowly push boundaries on what type of language they can ‘get away with’ before being moderated.
When discussing moderation, a popular theory that circulates is called the broken windows theory. This theory expresses that if there are signs of antisocial behavior, civil unrest and disorder, as well as visible signs of crimes in the area, that it encourages further anti-social behavior and crime. Similarly, if you create an environment in which toxic and hateful behavior is common, the cycle will perpetuate into further toxicity and hatefulness.
One additional component not included in the table is the effects of implementing a verification gate. The ramifications of a verification gate are difficult to quantify and not easily summarized. Verification gates make it harder for people to join in the conversation of your server, but in exchange help protect your community from trolls, spam bots, those unable to read your server’s language, or other low intent users. This can make administration and moderation of your server much easier. You’ll also see that the percent of people that visit more than 3 channels increases as they explore the server and follow verification instructions, and that percent talked may increase if people need to type a verification command.
However, in exchange you can expect to see server leaves increase. In addition, total engagement on your other channels may grow at a slower pace. User retention will decrease as well. Furthermore, this will complicate the interpretation of your welcome screen metrics, as the welcome screen will need to be used to help people primarily follow the verification process as opposed to visiting many channels in your server. There is also no guarantee that people who send a message after clicking to read the verification instructions successfully verified. In order to measure the efficacy of your verification system, you may need to use a custom solution to measure the proportion of people that pass or fail verification.
‘Bad-faith’ content is a term that describes behavior done intentionally to cause mischief, drama, or toxicity to a community. They are also commonly referred to as bad actors, and are the type of people that should be swiftly dealt with and addressed directly.
‘Good-faith’ content is a term that describes user behavior with good intentions. When users are a positive foundation in your community, the members that join and interact with the established community will grow to adapt and speak in a way that continues the positive environment that has been fostered and established. It’s important to note that while ‘good-faith’ users are generally positive people, it is possible for them to state wrong or sometimes even harmful words. The importance of this distinction is that these users can be educated from their mistakes and adapt to the behavior you expect of them.
When users toe the line, they are not acting within good faith. As moderators, you should be directly involved enough to determine what is bad-faith content and remove it. On the other hand, education is important in the community sphere for long term growth. While you can focus on removing bad behavior from bad-faith users, reform in good-faith community members who are uneducated in harmful rhetoric should also be a primary goal when crafting your community. When interacting in your community, if you see harmful rhetoric or a harmful stereotype, step back and meaningfully think about the implications of leaving content up in channels that use this kind of language. Does it:
A core way to handle all de-escalation stands in your approach. Users, when heated up during a frustrating or toxic discussion, are easy to set off or to accidentally escalate to more toxicity. The key is to type calmly, and to make sure with whatever manner you approach someone to de-escalate, you do it in a way that is understood to be for the benefit of everyone involved.
Creating a healthy community that leaves a lasting, positive impact in its members is difficult. Moderators have to be aware, educated, and always on the lookout for things they can improve on. By taking the initiative on this front, your community can grow into a positive, welcoming place for all people, regardless of their race, gender, gender identity, or sexual orientation.
Below are resources for further research and discussion on different types of slurs, symbols, and hate speech not referenced explicitly in this document.
Take the Discord Moderator Exam!Take the Exam