initial verification check dm logic

This commit is contained in:
William Kray
2025-04-06 21:07:01 -07:00
parent 49ae7bd66d
commit 5b973920ec
4 changed files with 177 additions and 6 deletions
+20
View File
@@ -148,6 +148,26 @@ please keep in mind that wordlist-based censorship is problematic and may redact
algorithm that is perfect is impossible. consider configuring your community such that censorship need only be applied
in a limited subset of rooms.
# user verification
configure your rooms (all, or a list of room-ids) to use the `check_if_human` setting. use this in conjunction with a room power-level configuration that
requires elevated permission to send messages. for example, a "waiting-room"
with a default power level of -1 for new users, while the power-level required
to send messages in that room remains 0.
enabling this and associated configuration will perform the following
validation:
1. when a user joins one of these rooms, the bot will check to see if they have
permission to send messages.
2. if not, the bot will start a DM with that user and ask them to repeat a phrase,
randomly chosen from your list of verification phrases. they have three tries.
3. when they send the matching verification phrase, the bot will bump their power
level up to that required to send messages in your room, and leave the DM.
not the most user-friendly experience, but may help cut down if you are experiencing
significant spam in your rooms.
# installation
install this like any other maubot plugin: zip the contents of this repo into a file and upload via the web interface,