How to set up an automatic moderation bot on your MU Online community's Discord

Discord has become the nerve center of practically every MU Online community — it's where event announcements, ticket support, item trading, and inter-guild discussions all happen. As the community grows, manual moderation (only humans reading every message) simply doesn't scale: spam, phishing links disguised as "free item websites," and message flooding happen faster than any team can react. A well-configured automatic moderation bot handles the obvious volume and frees up human moderators for decisions that require judgment. This tutorial shows how to set up this automation layer, from basic anti-spam rules to integration with the staff audit log.

Why automate part of moderation

Consider a MU server with 3,000 active Discord members: even with 5 moderators on shift, it's impossible to read every message in real time. A moderation bot covers what's objectively detectable by rule — unauthorized links, banned words, repeated message flooding, mass mentions — freeing up the human moderator for cases that require context: an escalated discussion, a report of a scam between players, a permanent ban decision. Automation doesn't replace human judgment; it reduces the volume that reaches it.

Choosing the foundation: native AutoMod vs. third-party bots

Discord has a native system called AutoMod, free and built-in, plus established third-party bots like Carl-bot, Dyno, and MEE6.

For most MU servers, the combination of native AutoMod (basic rules) + Carl-bot or Dyno (extra logging and warning features) covers the essentials at no cost. Larger communities with recurring game-specific scam patterns tend to move toward a custom bot.

Essential automatic filter rules

Set up, at minimum, these filter categories before considering automatic moderation "ready":

Configuring native AutoMod (step by step)

1. Go to Server Settings → Moderation → AutoMod.
2. Create a Custom word filter rule, adding terms specific to the MU Online context: known scams ("upfront payment," "selling banned account"), common offenses in Portuguese.
3. Enable the Mention spam filter, limiting how many mentions a message can contain before being blocked.
4. Set up the Malicious link filter (Discord maintains its own up-to-date list of suspicious domains) — keep it enabled by default.
5. Define the automatic action: block message, apply timeout, and send an alert to a specific staff log channel.
6. Create per-channel exceptions — for example, the item trading channel may need to allow Imgur screenshot links, so add that channel to the link filter's exception list, if applicable.

Progressive warning system

Banning on the first offense (except for serious cases like confirmed phishing) creates unfairness and reduces the server's active population. A progressive warning model, configurable in most moderation bots:

Serious cases — confirmed phishing, an attempted scam with concrete proof, threats — should skip straight to manual review and an immediate ban, without waiting through the full progressive-warning cycle.

Join verification (raid prevention)

Raid attacks (hundreds of fake accounts joining at once to flood the server) are common in popular gaming communities. Set up:

• Server verification level set to "High" or "Highest" (Settings → Moderation → Verification Level), which requires a minimum Discord account age and email/phone verification before sending messages.
• Welcome screen with mandatory rules (Settings → Community), requiring members to accept the rules before accessing channels.
• Abnormal growth alert — many security bots (e.g., Wick, Security bot) detect join spikes in a short interval and can automatically lock new joins until manual review.

Integrating the bot log with the GM command log

A practice that greatly improves audit quality: centralize the moderation bot's automated action log in the same channel (or a sibling channel with cross-references) where manual GM actions from the staff evaluation process are recorded. This lets you reconstruct a case's complete history — for example, a player banned in-game for duplicating items might have, on Discord, a history of warnings for attempting to sell items outside the rules, reinforcing the context of the decision.

Example log channel structure:
#log-automod       → automatic bot actions (spam, link, banned word)
#log-staff-gm       → in-game GM commands (ban, item, teleport)
#log-decisoes-finais → consolidated summary of cases resulting in a ban/severe punishment

Configuring exceptions for trade channels

Trading and item-sale channels (common on MU servers, even those that prohibit real-money RMT) need adjusted rules — a generic link filter can block legitimate inventory screenshots hosted on Imgur or the Discord CDN. Create a specific exception list for these channels, but keep the filter active for words related to RMT (real-money trading), since this usually violates both Discord's terms and the server's own rules.

Periodic review of AutoMod rules

Scams and spam patterns evolve — what worked 6 months ago may have a new variation (e.g., a scammer writing "P i x" with spaces to dodge the exact-word filter). Reserve a monthly review of AutoMod rules:

• Review the log of automatically deleted messages, looking for false positives (legitimate messages blocked by mistake).
• Update the banned words/expressions list with new variations identified by the staff.
• Adjust spam/flood thresholds if the server's normal activity has changed (a big event generates more legitimate messages per minute).

Common errors and fixes

Automatic moderation checklist

• Native AutoMod or third-party bot configured with essential filters (spam, link, banned words).
• Progressive warning system defined, escalating before a ban.
• Join verification raised to prevent fake-account raids.
• Exceptions configured for legitimate trade channels.
• AutoMod log centralized or cross-referenced with the GM command log.
• Monthly review of filter rules to cover new scam/spam patterns.
• Serious cases (phishing, threats) configured to escalate directly to manual review.

With automatic Discord moderation running alongside human staff, the natural next step is reviewing how that same team is evaluated and organized within the game — see the MU Online server creation tutorial to understand how to connect Discord community management with the server's overall operation.

Frequently asked questions