Move the typers template out from the template modification and <xf:include> it as a regular template.

This will make it easier to edit :)

FingerprintJS seems to produce false reports 80% of the time.
Can we integrate shared IP detection into it?

What I mean is that I only flag those users when both FingerprintJS and IP are shared.
Users with shared IPs have a 60–70% chance of being the same person.

As title - add support for conversation message alerts and auto updating of the counter in the header.

I know you've just released your chat pack, but this is an idea for a more light weight add on.

Basically just the "real time threads" addon, but for conversations:

1. User is typing block
2. Live messages
3. Live reactions

No need for new designs or popup windows - just leave conversations as they are with the added live bits.

You could then charge a smaller amount for this to catch the people that don't want the full chat pack :)

In the current version, permissions are set for user groups that have the right to use the bot.

I suggest adding another permission choice for a group (or groups) of users who don't have access rights, but to whom the bot responds by giving an announcement.

The add-on's settings could have a text field where the administrator could write a notice. Of course, it would be good to consider the language choices. Perhaps it would be easier to use phrases according to the languages, that would also be a perfectly functional solution.

The message would say that, unfortunately, you do not have sufficient access rights to use the bot. However, you can get them by getting a user upgrade and you can do that here <link>.

The bot would not react in any way to user groups that were not selected to receive a notification.

Issue
The AI Bots respond so quickly that it breaks immersion for forum users. When a user taps “submit” the bot already has a two paragraph response to their thread by time their page refreshes. It’s unnatural.

Solution
An option to delay responses to threads and posts by an admin selected timeframe. It could be set up like a scheduled post. When the response is sent back from ChatGPT to Xenforo a post can be scheduled for X minutes after [current time].

This will allow for more natural and less spammy feeling bots.

It could be great if this addon could search for threads with less than x words in the title. For example, I want to run this against all threads with 2 words or less in the title, as these are the low-quality ones.

The "Switch Account" plugin is great for administrators and moderators who want to easily switch between their accounts, without having to log out and enter multiple passwords each time.
At the same time, currently after logging out of the user account, it will be necessary to log in again to all duplicate accounts and enter all their passwords.
I would like to add a feature that will save the duplicate accounts in the site's memory, so that even after disconnecting, when you reconnect to one of the duplicate accounts, you will have the option to switch between the other accounts without having to enter passwords.
Thanks.

It would be nice if we could add a blacklist to the ChatGPT Title Corrector plugin and add things to this blacklist that we don't want suggested in the title. Some punctuation marks can also be annoying, optional punctuation marks can also be turned off.

I'd like to suggest an enhancement to the existing Auto-Title add-on. It would be great if we could add a custom prompt when creating titles, such as incorporating emojis into the titles. Emojis are valid characters, and Google displays them in search results. A title with an emoji at the beginning could be more eye-catching. Imagine hundreds of topics with engaging emojis in their titles. However, with the current Auto-Title add-on, we can't request the inclusion of emojis. Adding this feature would improve the appeal and visibility of our forum topics.

I would love to see Off the record messaging added to the jabber notifications plugin. I loved using it, although it was quite slow.

Suggestion to make new addon Ethereum payment. Like litecoin payment and bitcoin payment
Working with Ethereum and with main ERC20 tokens: USDT, USDC BUSD, DAI
Daemon may be geth - https://geth.ethereum.org/docs/getting-started/installing-geth
JSON-RPC Server - https://geth.ethereum.org/docs/interacting-with-geth/rpc

the multi-account detection system works very well! However, I have a few requests to make for potential features.

1. Proper handling of banned members

Having the add-on actually give out infractions or have the option to enable this would be great!

Click to expand...

2. Bypass permission for chosen user groups

While developing, occasionally I'll need to login and test different feature sets. Having the ability to bypass the detection methods, or choosing which ones would be a lifesaver!

Click to expand...

Thanks for your dedication towards these great add-ons. I love your work and hope you continue to make many more!

It would be good for the reply limit to be on a per forum basis, so that some forums might have just one response, other could be more.

I have found that when there are further responses via the bot being quoted this is less useful in serious info forums because users pile on and start arguing with the bot. ("Bot baiting"). This happens especially if the answer is wrong in which case I find it better to just have a mod intervene and point out the misinformation because other wise the bot starts hallucinating new answers

However this behaviour may be more useful in more more OT or fun forums

Dear @021 and all,

Have a nice day :)

I would like to add a private chat functionality for direct Conversations with the ChatGPT bot.

And members on my forum can discuss directly with ChatGPT bot via conversation.

Thank you,

Maybe belongs in the framework addon but some sort of budget management or throttle management i.e. max number of ChatGPT transactions per day/month etc. per user.

Global maximum ceiling too. In case you get spammed or some job goes out of control. I know you can manage some things on openai but this allows for granular controls.

I've seen a few posts about this, so I'm doing a more formal info / request.

PLEASE NOTE GPT-4 IS NOT AVAILABLE YET!!!

You can read all the release info here: https://openai.com/research/gpt-4

It clearly says you MUST get on their waitlist (which I did right when it became available).

"We are releasing GPT-4’s text input capability via ChatGPT and the API (with a waitlist). "

Their announcement gives good info about the API integration once a user is approved via waitlist:

"API
To get access to the GPT-4 API (which uses the same ChatCompletions API as gpt-3.5-turbo), please sign up for our waitlist. We will start inviting some developers today, and scale up gradually to balance capacity with demand. If you are a researcher studying the societal impact of AI or AI alignment issues, you can also apply for subsidized access via our Researcher Access Program.
Once you have access, you can make text-only requests to the gpt-4 model (image inputs are still in limited alpha), which we will automatically update to our recommended stable model as we make new versions over time (you can pin the current version by calling gpt-4-0314, which we’ll support until June 14). Pricing is $0.03 per 1k prompt tokens and $0.06 per 1k completion tokens. Default rate limits are 40k tokens per minute and 200 requests per minute."

Note that GPT-4 is MORE expensive than the current 3.5 Turbo option.

In creating this suggestion thread I'm hoping that once I'm approved for GPT-4 that I'll be able to toggle / enable this newer model.

It's great to have the replies field either yes or no default. But would it be possible to have it unticked by default, and the option to make it required or not.

This is because whatever the yes/no option is set as default some users just don't see it. So many may even miss the opportunity to get the post answered by bot.

So if they can't post until they tick one or the other then they will definitely know they can get a bot answer if they want.

I have GPT 3.5 integrated in Obsidian and can within any note get contextual information on what I am writing in the whole note or in a line or paragraph.

For our forum bots or autoresponders or even contextual help with answers to posts make not much sense.

What I would like to see is the possibility to work with text that I myself put into the editor. That could be information on what I am writing, help with arguments on my standpoint, a summary of a longer text that would otherwise be a copyright problem, or a translation of a foreign language text right made inside the post, instead of copy-pasting back and forth between (e.g.) DeepL and the forum.