Skip to main content
Content Control

Advanced
Content Filters

Fine-grained control over what content is allowed in your group. Links, media, keywords, and more.

  • Instant filtering
  • Regex support
  • Whitelist/blacklist
Incoming messagesFilter layersClean chat

Filter Types

Link Filters

Control which links can be shared.

  • Block all links
  • Whitelist specific domains
  • Blacklist specific URLs
  • Allow only trusted members

Media Filters

Control photos, videos, and files.

  • Block photos
  • Block videos
  • Block stickers/GIFs
  • Block documents

Keyword Filters

Block specific words and phrases.

  • Word blacklist
  • Phrase blocking
  • Regex patterns
  • Case sensitivity options

Mention Filters

Control @mentions and forwards.

  • Block @mentions
  • Block forwards
  • Block channel links
  • Allow admins only

Filter Actions

Delete
Remove message
Warn
Issue warning
Mute
Silence user
Kick
Remove user
Ban
Permanent ban
Log
Record only
Replace
Censor text
Notify
Alert admins

Advanced Features

Regex pattern matching
User role exceptions
Time-based rules
Message length limits
Cooldown periods
Stricter for first messages
Import/export rules
Per-filter statistics
Test mode
Cascade rules
API integration
Bulk rule management

Example Rule

Filter: Links
Mode: Whitelist
Allowed: youtube.com, github.com
Action: Delete + Warn
Except: Admins, trusted users
Message: "Links are restricted here"

Why Use Filters?

  • Prevent Spam - Stop unwanted content before it's seen.
  • Keep Focus - Keep discussions on-topic.
  • Protect Members - Block malicious links and content.
  • Complement AI - Add custom rules to enhance AI detection.

How filter rules match and act

Filters are deterministic rules you define, checked against every message as it arrives. You choose what to match — links by domain whitelist or blacklist, media by type, keywords as plain words, phrases, or full regex patterns — and what happens when it matches, from a silent log entry through delete, warn, mute, kick or ban. Rules carry exceptions, so admins and trusted members can be exempt, and can be scoped by time, message length or how new the sender is, letting you apply stricter handling to a member's first messages than to a regular's. A test mode lets you try a rule before it acts.

Because filters do exactly and only what you wrote, they are predictable in a way AI cannot be — a link to a competitor, a banned word, an unwanted file type is caught the same way every time, and per-filter statistics show what each rule is actually catching. That precision is also the limit: a rule matches the literal pattern, so spammers who reword, use look-alike characters or hide text in an image slip a naive keyword list. This is why filters complement AI rather than replace it — you encode the rules specific to your community, and the AI pipeline covers the disguised cases your patterns were never written to see.

Frequently asked questions

What can I filter?

Keywords and phrases, links and domains, forwarded messages, media types (photos, videos, stickers, voice messages, documents), mentions, and Telegram usernames — each with its own action, from silent delete to mute or ban.

Do filters support patterns, not just exact words?

Yes. Filters match word variations, spacing tricks, and character substitutions that spammers use to dodge naive keyword lists — so "f r e e c r y p t o" doesn't slip past a "free crypto" rule.

Can I make exceptions for admins or trusted members?

Yes. Admins are exempt by default, and you can whitelist specific members, domains, or link destinations so your own resources are never flagged.

Where teams use this

Real community types where this feature does the heavy lifting.

Ready to Get Started?

Start protecting your community today. Free plan available.