Grant an exception for certain bots from hellbans
This is similar to https://gitlab.com/gitlab-org/gitter/webapp/issues/1958.
As per the conversation in gitterHQ/gitter, bots can be banned across Gitter if enough people report on them. This opens up a potential abuse vector as as far as I can tell, there is no tracking to see if these reports are legitimate (https://gitlab.com/gitlab-org/gitter/webapp/blob/2bad5c88524c884b2b3e67cf5a0f82a15173fa29/modules/chat-reports/index.js#L150).
The threshold is 5 messages currently, which means if 5 people file reports against a bot then the bot is hellbanned. I would petition that Gitter do not ban (some) bots based on reports as it will disable bot functionality for the whole network.
I'm currently maintaining @matrixbot (matrixbot
) on behalf of the matrix.org foundation, and regularly the bot is reported and this means the whole bridge is knocked out until I ask the Gitter network admins to clear the ban. We bridge over 1000 Matrix rooms into the Gitter network, and so the impact is extreme.
#1958 would not be a workable solution to this problem, as it means that the bot would need to become admin for all the rooms it is joined to. I do not believe Gitter users would want to grant permissions to the bot user everywhere.
As an aside, we are not able to tell which rooms we are being reported from and this makes it challenging to address any genuine concerns of uses.