Replies: 4 comments 10 replies
-
isn't this what https://github.com/umputun/tg-spam?tab=readme-ov-file#training-the-bot-on-a-live-system-safely
|
Beta Was this translation helpful? Give feedback.
-
I'm adding |
Beta Was this translation helpful? Give feedback.
-
I partially checked today the new work scheme (at first without a fix - I thought I had configured something wrong, then I saw the v1.12.1 update) - this is what I expected, except for one thing - I expected that after confirmation through the administrators chat user will move from the “exceptions” (users with custom permissions) to the “blacklist” and will be (automatically) removed from the chat. |
Beta Was this translation helpful? Give feedback.
-
In general, as far as I understand, when a bot detects spam, it bans the user by blacklisting it (a standard Telegram feature)
But sometimes the bot mistakenly bans an ordinary new user, then you have to write a personal message and return the user manually, because often the privacy settings prohibit adding to groups.
The suggestion is the following - delete messages and limit the rights of suspicious users (rather than blacklisting them) if an admin chat is configured. And let the administrators blacklist them by confirming the ban.
If the chat for administrators is not configured - leave the current behavior - ban from the first message.
Beta Was this translation helpful? Give feedback.
All reactions