Telegram Group Moderation

Moderate Telegram groups with a bot by receiving message/webhook events, extracting text/caption/media context, applying anti-advertising and anti-contact po...

This package is an integration skeleton that will forward Telegram updates and message content to an external moderation endpoint and perform Telegram API actions using a bot token. Before installing or enabling it: 1) Confirm the source repository and review the example scripts yourself. 2) Do not supply real production secrets until you audit the code. The examples require TELEGRAM_BOT_TOKEN and MODERATION_CORE_ENDPOINT (and optionally MODERATION_CORE_TOKEN, TELEGRAM_WEBHOOK_SECRET, etc.), but the registry metadata doesn't declare these — expect to set them as environment variables. 3) Ensure the moderation endpoint is trusted (you are intentionally sending message text, chat_id, user_id, and related metadata). 4) Use a test Telegram bot and a test group with dry-run enabled first; restrict allowed_chat_ids and allowed hosts in configuration. 5) If you plan to run this in production, prefer storing tokens in a secrets manager, enable webhook secret verification, and move offense storage to Redis/DB rather than local files. If you need higher assurance, ask the publisher to update the package metadata to declare required env vars and runtime binaries and to provide a release provenance (signed release or known GitHub release) before deployment.

Type: OpenClaw Skill Name: telegram-group-moderation Version: 0.9.0 The telegram-group-moderation skill bundle is a well-structured integration layer for connecting Telegram bots to a moderation policy engine. It provides clear architectural guidance and multi-language implementation examples (Python, Go, Java) for handling webhooks, normalizing message data, and executing moderation actions like deleting messages or banning users. The code follows security best practices by utilizing environment variables for sensitive tokens (TELEGRAM_BOT_TOKEN) and implementing webhook secret validation, with no evidence of malicious intent, data exfiltration, or harmful prompt injection.