Security dashboard
Anti-nuke thresholds, raid containment, spam and link filtering, member verification, and the cross-server Trust Network — all configured from a single tabbed page with a live overview.
Before you start
- You are an admin or have the security.manage capability on this server
- SYNTHET has Manage Server, Manage Roles, Manage Channels, Ban and Kick permissions
- You have a private staff channel ready to receive security alerts
- You know which roles count as moderators or trusted members
What you'll see
The page header shows the server name, a Refresh button and a Save button. Below it, four tabs switch between read-only monitoring and the full settings surface.
Overview tab
Read-only dashboard. Use it to confirm protections are running and to spot threats from the last 24 hours at a glance.
Status tiles
A six-tile strip shows the live state of each protection:
- Anti-nuke — Active or Disabled.
- Anti-raid — shows the active sensitivity profile (Relaxed, Default, Hardened or Lockdown-ready) when on.
- Anti-spam — Active or Disabled.
- Anti-link — Active or Disabled.
- Member Verification — shows the template name (Fast-friendly, Balanced, Raid-hardened) when on.
- Trust Network — link to view or configure the cross-server reputation feed.
Threat counters and charts
- Four stat cards show recent threats split by severity: Critical, High and Medium/Low.
- A 24-hour bar timeline plots threats per hour. Tall bars at unusual times are usually worth investigating.
- A risk-distribution donut breaks decisions into Clean, Watch, Quarantine and Block buckets so you can see what your filters are catching.
- A Recent threats table lists the latest events with timestamp, type, severity chip and description.
Settings tab — Anti-nuke
Catches mass-destructive admin actions like channel deletion sprees, role wipes and webhook spam. Triggering breaches lock down the offending actor.
Open DashboardSecuritySettings.
Master toggle
The Anti-nuke protection toggle at the top of the card enables every threshold. When it's off, nothing else in this card runs.
Detection thresholds
An
| Setting | Type | Default | Description |
|---|---|---|---|
| Channel deletes | Number | 5 | Channels removed by a single actor in 60s. |
| Role deletes | Number | 3 | Roles removed in 60s. |
| Permission overwrites | Number | 10 | Mass overwrite changes in 60s. |
| Admin escalations | Number | 2 | Admin perms granted to a non-admin in 60s. |
| Role rank height | Number | 10 | How high a created role can sit before flagging. |
| Webhook create spike | Number | 5 | Webhooks created in 60s. |
| Mass bans | Number | 5 | Bans issued in 60s. |
| Mass kicks | Number | 5 | Kicks issued in 60s. |
How long the offending actor stays locked out after triggering anti-nuke. 5–1440 minutes.
Example: 30
Layer the Behavioural Insights Engine on top of the static thresholds for smarter pattern detection. Recommended on busy servers.
Users who can never trigger anti-nuke. Server owner is always implicitly whitelisted.
Channel where anti-nuke alerts post. Leave empty to fall back to the global Security log channel.
Roles to ping when an alert fires. Use sparingly — only roles that should wake up at 03:00.
Settings tab — Anti-raid
Detects coordinated joins and mass disruption. Reacts with quarantine, lockdown or permission freeze.
Sensitivity profile
Pick one of four presets from the Sensitivity profile dropdown. The card on the right shows the resolved thresholds for each preset.
- Relaxed
- High tolerance — only obvious raids fire. Use on small or mostly-private servers.
- Default
- Balanced for most public servers. Recommended starting point.
- Hardened
- Lower thresholds, faster reactions. Use if you've recently been raided.
- Lockdown-ready
- Aggressive triggers, quarantine on first sign of trouble.
Containment actions
- Quarantine — assigns the quarantine role to new joiners during a raid.
- Lockdown — escalates to a server-wide channel lockdown for catastrophic events.
- Freeze permissions — temporarily strips role permissions until the incident clears.
- Auto-cleanup — removes the quarantine role from caught joiners once the incident ends.
Use the Behavioural Insights Engine to spot subtler raid patterns.
Role new members receive during an active raid. Auto-created on first enable if left blank.
Where raid alerts post. Falls back to the Security log channel.
Roles pinged when a raid is detected.
Settings tab — Anti-spam
Stops message floods, mention spam, oversized messages and duplicate-message bursts.
Detection thresholds
| Setting | Type | Default | Description |
|---|---|---|---|
| Message rate limit | Number | 5 msgs | Max messages per user inside the time window. |
| Time window (seconds) | Number | 10s | Sliding window for the rate limit. |
| Max mentions | Number | 10 | Max user/role mentions in a single message. |
| Max newlines | Number | 15 | Max line breaks in a single message. |
| Duplicate threshold | Number | 3 | Identical messages before a duplicate flag. |
Choose what happens when spam fires: delete message, timeout 1/5/10 minutes, or kick.
Example: timeout5
Roles that bypass spam detection (mods, bots).
Channels where spam detection is off (memes, spam-zone).
Settings tab — Anti-link
Filters URLs and Discord invites with a domain whitelist.
What happens to a blocked link: delete, warn, timeout 1 or 5 minutes, or kick.
Example: delete
Also block discord.gg / discord.com/invite links, even if the URL host isn't blacklisted.
Domains that are always allowed. Add one per line.
Example: github.com
Roles that can post any link freely.
Channels where links are always allowed (e.g. #news, #links).
Settings tab — Member Verification
Optional gate for new joiners. Risk-scored challenges keep humans in and bots out.
Template preset
Pick a template from the dropdown to set sensible defaults across challenges and risk weights:
- Fast-friendly
- Click-verify only, low friction. Good for established communities.
- Balanced
- Default. Mix of click-verify, rules acknowledgement and time delay.
- Raid-hardened
- Stricter — adds simple/puzzle pick and manual review for high-risk joiners.
Challenge per risk tier
Each risk tier (low / medium / high) gets its own challenge from this list:
- CLICK_VERIFY — single button click.
- RULES_ACK — read and accept server rules.
- SIMPLE_PICK — pick the matching emoji or option.
- PUZZLE_PICK — slightly harder picture/text puzzle.
- TIME_DELAY — wait N seconds before the button works.
- MANUAL_REVIEW — a human moderator approves.
Role applied while waiting to verify.
Role granted after a successful challenge.
Channel where the verification panel message lives.
Where verification outcomes are logged.
Where manual-review requests are routed.
Skip the challenge entirely for the cleanest joiners.
Failed attempts before the user is escalated to manual review.
Example: 3
Waiting period after a failure.
Example: 60
Roles that skip verification entirely (rare — usually only staff).
Trust Network tab
PRO feature. Subscribes this server to a shared moderator decision feed so confirmed bad actors are recognised on arrival.
What this tab shows
- A short description of the network and how decisions propagate.
- A primary "Open Trust Network" button that opens the dedicated page.
- A "Recent decisions" panel listing the last 24 hours of inbound decisions, or an empty state if you haven't enabled the feed yet.
Audit tab
Read-only timeline of every change made on this page. Use it to answer 'who lowered the anti-nuke thresholds last week?'
What gets logged
- Toggling any protection on or off.
- Changing thresholds, sensitivity profiles or actions.
- Adding or removing whitelist entries, exempt roles and exempt channels.
- Verification template switches and challenge changes.
Common tasks
Turn on the recommended baseline
The safe minimum for a public server.
- Open DashboardSecuritySettings.
- Enable Anti-nuke and leave the default thresholds.
- Enable Anti-raid with the Default profile.
- Enable Anti-spam with action set to timeout5.
- Enable Anti-link with Block Discord invites on.
- Click Save in the page header.
Add staff to whitelists and exempts
Stop your own moderators from triggering filters.
- Open the anti-nuke card and add admins/owner-tier users to Whitelisted users.
- In anti-spam, add your moderator role to Exempt roles.
- In anti-link, add the staff role to Exempt roles and add #staff to Exempt channels.
- Save the page.
Switch on member verification
Stop drive-by spam accounts before they post.
- In the Verification card, toggle Member verification on.
- Pick Balanced as the template.
- Set Verified role to whatever role currently grants channel access.
- Pick a Panel channel visible to unverified members and a private Staff channel for manual-review escalations.
- Save and post the verification panel from the dashboard.