Guide/Protect your server/Protections overview
Protections overview — Discord Bot Guide
The map of every protection, the shared list of punishments, alerts, and the set-anything-by-path command.
Quick answer
Read and change anything by path
!security paths
!security get antinuke
!security set antinuke.thresholds.channelDelete.limit 3
!security set cleanup.hours 2Ask the guide
Step by step
Security is one engine with several protections. Each can be on or off, each has its own thresholds and punishment, and all of them write to the same incident and case system. !security status shows every protection, its metrics for the last 24 hours and recent threats; the dashboard page Security → Protection is the full editor with the same numbers.
| Protection | What it watches | Command |
|---|---|---|
antinuke | Every destructive action — bans, kicks, timeouts, purges, channel/role/webhook/emoji create-update-delete, dangerous permission or role grants, bot adds, prunes, server changes. Strict mode bans a non-whitelisted actor on the FIRST action; permission grants say which role may do exactly what | !antinuke · !antinuke permrole |
Discord AutoMod | Native edge rules mirrored from your message settings, so spam is blocked before it posts | !security automod sync |
antiraid | Join floods (N joins in M seconds) and a risk score per joining account (age, avatar, name, invite used) | !antiraid |
join gate | Individual rules on join: no avatar, account too young, unauthorised bot, invite link in name, username pattern | !antiraid joingate |
automod | Spam, mass mentions, links and invites, blocked words, ghost pings, bad nicknames | !automod |
heat | A score per member that grows with risky behaviour and decays every minute; thresholds escalate warn → delete → timeout → quarantine → ban | !automod heat / threshold |
verification | New members must verify (button or web) before they can see the server | !verification |
moderator limits | Caps how many bans/kicks/timeouts/… any moderator can do per window, whitelisted or not | Dashboard → Protection → Moderator limits |
lockdown / panic | Lock channels for a duration; panic mode locks everything, blocks joins and snapshots the server | !lock · !lockall · !panic |
backups | Snapshots of roles, channels, permissions, emojis, the security configuration and (optionally) messages; diff-based restore; send stored messages back | !backup |
Punishments
Every protection chooses from the same actions: none, alert, warn, delete, timeout, kick, ban, quarantine, lockdown, removeRoles. Each action creates a numbered case, an incident when it is part of a wider event, a log entry, and an alert when severity reaches the alert threshold.
Alerts
!security alerts #security-alerts @Staff high channel, role to mention, minimum severity (info/low/medium/high/critical)
!security webhook add https://… raid.detected,nuke.detected
!security webhook list · !security webhook remove <id>Everything by path
Every single setting is reachable with !security set <path> <value>; !security paths lists them and !security get antiraid prints a whole section. Values: on/off, numbers, ids, none. Example: !security set antinuke.thresholds.channelDelete.limit 3.
Copy and try
Read and change anything by path
!security paths
!security get antinuke
!security set antinuke.thresholds.channelDelete.limit 3
!security set cleanup.hours 2Good to know
- The Protection page starts with a live strip: current state (normal / level / RAID / PANIC), joins in the raid window against the limit, active lockdowns and quarantined members.
- Structured lists (permission grants, heat ladder) have their own editors; !security set refuses to overwrite them.
- Every write is validated: a wrong type or an unknown id is refused with a reason.
Still stuck? Ask in the support server.