Telegram Toxic Language Detector with AI Moderation

This workflow automates moderation in Telegram chats by triggering on new or edited messages. It uses the Google Perspective API to analyze message text for toxic attributes like identity attacks, threats, and profanity, scoring them on a 0-1 probability scale.

If the toxicity score exceeds 0.7, it sends an automated warning message ('I don't tolerate toxic language') back to the chat. Otherwise, it takes no action via a NoOp node. This ensures clean, safe conversations without manual intervention.

Benefits include real-time moderation, reduced admin workload, and scalable protection for communities, support groups, or channels. Ideal for Discord alternatives, customer service bots, or public forums where toxicity can harm engagement.

Use cases: Community managers monitoring large groups, businesses maintaining professional Telegram channels, or educators fostering positive discussion spaces. Easy to set up with Telegram bot credentials and Perspective API keys.