I'm running purple-discord and Ripcord simultaneously, and noticed that the same message was displayed with different authors. While purple-discord (and official web app as well) displays the authors correctly, in Ripcord all messages that come from the gate are displayed with the same author. This author has #0000 tag, if important, and all messages that the gate sends have #0000 tag too, but the usernames are correct.

Ripcord for Discord does not respect the "fake user" data that bot messages can be sent with. It simply displays the newest information for the bot user that it has received from the user, instead of assigning it per-message.

This is a big change I would have to spend a while implementing, and I don't have time for it right now. It's also a feature that can easily be abused by bots, so I'd have to be careful when implementing it to make sure it's clear to the user of Ripcord that the bot is forging per-message fake identities.

Please do not mark minor issues as Important.

This is a really old issue but I have been seeing this issue a bit recently. I do not remember the date that you implemented BOT tags into discord messages when it is implied, not something I cared to look up and if this ticket pre-dates the feature, but it is a step in the right direction.

You could in addition add a webhook identifier to the message, either by improving the tag or any other identifier. Each message sent contains a webhook_id element in the MESSAGE_CREATE object (These match the id of the author object), but it gives a reliable way to detect that messages are being sent from a webhook, rather than just looking at a zeroed discriminator. You can't reliably track webhooks from bots, but it's also not active bots in a guild that send these messages, as they are also imported from other discord channels through announcement type channels, through discord's administrative webhooks (see [system] public server announcements), but there is an unknown type one of my contacts use that are a self-run annoucement webhook with [server] tags attached to the message.

There are limitations on their side, but it shouldn't provide too much of a worry. The real gripe would be accustoming the software for the per-message identity. At least with Ripcord we can ignore messages from webhooks with their IDs, this is a feature that the web client actually lacks, therefore we have a bit more power already at our hands if there are users abusing this.