Moderation & Safety

AI catches what keyword filters can't

Context-aware spam detection, real-time raid alerts, and a layered defense stack give your admins the edge against coordinated attacks, without false-positive bans.

Start for freeBook a demo
Works across
AI spam detection

Smarter than a word list

Keyword filters break the moment spammers change a single letter. Fren One's AI reads full message context, detecting spam, scam patterns, and coordinated bot attacks even when no banned word appears. It learns from your community's conversation style, so false positives stay low.

Context-aware classification, not regex matchingFren One reads the whole message, not a word list, so rewrites and unicode tricks still get caught.
Detects coordinated bot attack waves before they flood the chatJoin velocity, message rate and account age are tracked together, and your configured lockdown fires on its own.
Works across Telegram and Discord from a single toggleConnect every community to one project and the same protection applies across all of them.
How it works

Detect, act, review, automatically

Every layer follows the same loop, so nothing slips through while your admins are offline.

  1. 01Detect

    AI classification, join-velocity, and message-rate monitoring watch every message and every join in real time.

  2. 02Act

    Your configured action fires on its own: delete, mute, slow mode, captcha gate, or a temporary lock.

  3. 03Review

    Every action lands in the project inbox with a full audit trail, ready for your team to review after.

Real-time defense

Automatic reflexes, always on

Stop raids before the damage is done

Join velocity, message rate and account age are watched together. When the pattern matches, your configured response fires on its own, and you get the alert either way.

End message floods without silencing conversation

One member bursting past your threshold gets muted or deleted. Everyone else keeps talking, and your slow-mode settings keep working alongside it.

Full protection stack

Every threat, one place

Layered defenses that complement each other. Toggle what you need, leave the rest off.

Impersonation defense

Detects members copying admin usernames or display names and flags or removes them automatically.

Word & user blacklists

Permanent pattern-based filters for slurs, links, and repeat offenders, synced across your communities.

Custom filters

Regex and keyword rules with per-filter actions: delete, warn, mute, or kick. No coding required.

Warns system

Issue warns that accumulate toward escalating consequences: mute, kick, or ban, configurable per community.

Member reports

Members flag rule-breaking content directly in chat. Reports queue in your dashboard, not your DMs.

Vote-ban

Let your community self-govern: enough votes against a member triggers a hold and admin review.

Raid detection

Join-velocity and message-burst analysis fires an automatic defensive action and pings your inbox.

Flood control

Per-member message rate limits with configurable mute or delete, separate from channel slow mode.

Notifications

Route moderation alerts to an admin channel, the project inbox, or both, with configurable severity.

AI smart triggers

Intent-based triggers that activate on meaning, catching frustrated members or off-topic tangents automatically.

FAQ

Common questions

What admins usually ask before switching AI moderation on.

Still unsure? Talk to us →See how Fren One compares →
No, they work together. Keyword filters catch obvious repeat offenders instantly; AI classification handles novel patterns that no keyword list would catch. You control which runs first and what action each triggers.
Thresholds are configurable and you choose the auto-action. The gentlest option is adding a captcha gate rather than locking, so real new members can still join while the spike is assessed.
Yes. Use as much or as little of the platform as you like. Everything is free during the open beta, and a free tier covering core moderation is planned to stay after launch.
Warns are logged centrally in the project dashboard, visible to every admin on your team. Any admin can review, escalate, or clear a warn, with a full audit trail of who did what.

Safe communities grow faster

Set up AI moderation in minutes. No developer required.

Start for freeBook a demo