TGtrigger

Freelancers and small agencies find clients in public Telegram chats, where a real request sits between service ads, self-promotion and small talk. Reading dozens of chats by hand every day does not scale.

A userbot connects your Telegram accounts and listens to the chats you pick. Triggers are plain phrases or regular expressions, scoped to one chat or to all of them. Each match then passes the schedule, ignored authors, negative filters and duplicate checks, and the same author is muted for 24 hours so one busy person cannot flood the feed. Obvious service ads are marked as spam first by a fast filter of 57 phrases and a price pattern, and a small AI model labels what is left as hot, warm, cold or spam and adds a one-line reason. If the AI is unavailable, the message passes as warm, so a lead is never lost to an outage. Spam is logged and never forwarded; real leads go to people, bots or channels in Telegram and to webhooks.

The dashboard is a single page that updates live: a bell with new leads, a sound for hot ones, charts by day and hour, top triggers, chats and senders, and CSV export.

The first version, on 24 February 2026, already had the monitoring and the dashboard, and in March the product got its own landing page. The service ran until April 2026 and was released as open source under the MIT license in June, so anyone can run it on their own server.

For the demo, the dashboard runs on fictional English chats and leads, a live feed adds a new match every few seconds, and every action that would reach Telegram is switched off.

Instead of reading every chat, you open a short list of messages that already passed your rules, each with an AI label and the reason behind it.

Code: github.com/A1exZabr/tgtrigger, MIT license.