Telegram Word Filter Bot: Block Inappropriate Content

Every online community faces the same challenge: keeping conversations civil without requiring moderators to read every single message. A word filter automates the first line of defense by automatically detecting and acting on messages that contain prohibited words or phrases. As part of a complete group management strategy, OmniGest includes a powerful, configurable word filter that works instantly.

How the Word Filter Works

OmniGest's word filter scans every message sent in your group against a list of prohibited words that you define. When a match is found, the bot takes the action you have configured -- deleting the message, warning the user, muting them, or a combination of responses.

The filter runs in real time, processing messages as they arrive. There is no delay or queue. The moment a message containing a blocked word appears, the bot acts.

Adding Words to the Filter

Adding words is straightforward. Use the word filter configuration through /config to manage your list. You can add individual words or phrases that you want blocked.

The filter supports:

• Single words -- Block a specific word regardless of context.
• Phrases -- Block multi-word phrases that should be caught as a unit.
• Patterns -- Use word boundaries to control whether partial matches are caught.

For complete configuration instructions, check the Word Filter documentation.

Word Boundary Detection

One of the most common problems with basic word filters is false positives. A naive filter blocking the word "ass" would also catch "class," "assistant," and "passage." OmniGest handles this with word boundary detection.

When word boundaries are enabled, the filter only matches whole words. The blocked term must appear as a standalone word, separated by spaces, punctuation, or the start or end of a message. This dramatically reduces false positives while still catching the content you want to block.

You can toggle word boundary detection on or off depending on your needs. For some terms, you want strict matching. For others -- like URL fragments or character sequences used to evade filters -- you might want substring matching.

Configurable Actions

Different groups need different responses when a filter match occurs. OmniGest lets you configure what happens when a blocked word is detected.

Delete Only

The message is removed, and no further action is taken. This is the least intrusive option. Users might not even realize their message was caught unless they check the chat. This works well for groups where occasional slip-ups happen but persistent violations are rare.

Delete and Warn

The message is removed, and the user receives a warning from the bot. The warning lets them know their message contained a blocked word and reminds them of the group rules. This approach educates users rather than punishing them silently.

Delete and Mute

For groups with strict content policies, the message is removed and the user is temporarily muted. The mute duration is configurable. This is effective for groups that have zero tolerance for certain types of language.

Escalation

OmniGest supports progressive enforcement. A user might receive a warning on the first offense, a short mute on the second, and a longer mute on the third. This graduated approach is fair to users who make honest mistakes while still dealing firmly with repeat offenders.

Use Cases

Profanity Filtering

The most straightforward use case. Add common profanity, slurs, and offensive terms to your filter. This keeps the group professional and welcoming, especially important for groups that include minors, customers, or professional contacts.

Anti-Spam Keywords

Spammers often use specific keywords like "earn money," "free crypto," "DM me for deals," or specific URLs. Adding these phrases to the word filter catches automated and manual spam alike. For a full multi-layer approach to spam defense, see our spam protection guide.

Competitor Mentions

Business or brand groups sometimes need to prevent promotion of competing products. Adding competitor brand names to the filter keeps the group focused on your product or community.

Political and Sensitive Topics

Groups that want to stay focused on their core topic can filter politically charged terms or hot-button phrases that tend to derail conversations. This keeps discussions productive without requiring moderators to manually steer every conversation.

Language Enforcement

Multilingual groups that want to enforce a specific language in certain discussions can filter common words from other languages. This is a blunt approach, but effective for groups with strict language policies.

Building Your Word List

Starting from scratch can feel overwhelming. Here are practical tips for building an effective word list.

Start Small

Begin with the terms that are most problematic in your group. Review recent messages that required moderation and identify the words that appear most frequently. Add those first.

Monitor and Adjust

Check the modlog regularly to see what the filter is catching. If you see false positives, adjust your word list or enable word boundaries for those terms. If you see violations slipping through, add the new terms.

Do Not Over-Filter

A filter that blocks too many common words makes the group unusable. Members become frustrated when innocent messages are deleted, and they leave. Aim for precision over coverage.

Consider Evasion

Persistent users will try to evade filters by using alternate spellings, adding spaces between letters, or substituting characters. You can add common evasion patterns to your filter, but recognize that a word filter is not a perfect solution. Combine it with AI moderation for more sophisticated content detection.

Get Started

The word filter is completely free with OmniGest. Add the bot, configure your word list, and let it handle content moderation around the clock.

Add @OmniGest_bot to your Telegram group and set up your word filter today. See the Getting Started guide for a complete setup walkthrough.