[XTR] AI Assistant

[XTR] AI Assistant [Paid] 1.0.35

No permission to buy ($45.00)
Overview FAQ Updates (34) Reviews (6) History Discussion
@Osman ,
Impressive tool. I haven't purchased it quite yet. But, I am inclined to.

Not sure if this suggestion has been made yet or if it is even popular.....I run a private site. As a thread starter, a user may not want the contents of the thread to be submitted to a third-party (public) LLMs and therefore, no summaries to be generated, no bot support for the thread, etc. Can the thread starter turn it off per thread?

Also, what protections of the tool are in place (or planned) to mitigate sharing PII (such as emails, birthdates, private conversations, etc.) with third party LLMs? Does the user have granular control over what is being shared with the service providers? Can users opt-in/out from having all of their data used? Does the admin have that control?

Is there (or can there be) an automated way to update the privacy policy page to link visitors/members to each provider/model privacy policy? This could help with shouldering liability.
Along those same lines of thought....Is there an overall visitor/member help page opt-in to notify members that AI tools are in use.....and force acceptance like the privacy policy?

Thanks!!
 
Hello @chris26237,

Thank you for your interest in our add-on. Your questions are very pertinent, especially for private communities, and we take these privacy concerns seriously. Here is a breakdown of our current features and upcoming plans:
  1. Per-Thread AI Control: Our tool already supports this functionality. A thread starter (or authorized moderators) can disable AI bot activity for a specific thread either during creation or later via the "Thread Tools" menu. When disabled, no content from that thread is processed by the AI, and bot interactions (summaries, replies, etc.) are restricted.
  2. PII (Personally Identifiable Information) Protection: In the current version, data is transmitted directly to the LLM providers. However, we are actively planning a "Data Masking" feature for an upcoming update. This will automatically redact sensitive information such as email addresses, phone numbers, and specific dates before they are sent to third-party LLMs.
  3. User Control and Opt-out: Administrators have granular control over which forum nodes have AI features enabled. Additionally, we are looking into implementing a global user preference that allows members to opt-out of having their content processed by AI tools across the entire board.
  4. Privacy Policy & Acceptance: While we don't currently have an automated link generator for privacy policies, we recommend updating your forum’s standard Privacy Policy and Terms of Service to include AI usage.
 
Osman updated [XTR] AI Assistant with a new update entry:

1.0.19

In this release, we focused heavily on User Privacy and Data Security. We know that running a private community means keeping your members' data safe and giving them control over how their content is used. With v1.0.19, we are introducing a powerful multi-layered privacy infrastructure.

What's New in this Release?
  • Global User Opt-Out: We believe in user autonomy. Your members can now go to their standard XenForo Account -> Privacy settings and check a...
Click to expand...

Read the rest of this update entry...
 
XFR00z0 said:
Members using this mod extensively, how much is average cost per month for tokens used in different GPT agents combined ?
How much additional you are paying ?
If you plan to make 200 new posts per day in 20 new topics, average of 500 words per post.
Click to expand...
I'm interested in this too.
 
@Osman if you can give a glimpse of this information, will be helpful.
As the costs can go very high to make a dying forum, go alive with help of AI (or until the costs come down)
 
I'd be interested in seeing some example threads that this modification/AI has created?
 
If you used Grok, it could cost you maybe a buck or two monthly, GPT a bit more, maybe $3 monthly... Claude Haiku costs me $5 last month, much higher for other Claude models (steer clear of Opus or Sonnet for general use. Grok and OpenAI are pretty cheap on API. If you were making 200 posts daily on API, then you're probably looking around $5 monthly depending on post size, prompt size, etc.
 
@Osman,

Great plugin and it works perfectly! However, I noticed a few issues on XenForo 2.3.9. The plugin version I am using is 1.0.19.
  1. Guest posts issue: If a thread contains posts by a guest, or if the thread itself was created by a guest, the generation fails with an error. This happens because a guest's username is evaluated as 0, even though the guest still has a specific display name. It would be great if the plugin could fetch the guest's name and send it via the API. Currently, all submissions involving guest posts are blocked due to this error.

    1779635028048.webp

  2. Mobile browser issue: On mobile Android using the Google Chrome browser, when generating a suggestion via the button, the generated result is not inserted into the reply area.
  3. Scheduled posting issue: For some reason, scheduled postings are not working. They are configured correctly and they do trigger, but the final results are never posted on the forum.

 
Another bug found. I created a new Suggestion Tone and configured my own custom prompt for it. However, when triggering this tone via the generation button, the custom prompt is not applied. Instead, a standard tone (like xentr_aia_tone_friendly or another default one) is randomly substituted. The API should be receiving my custom prompt instead.
 
Hi @Qwatt
Could you please create a support ticket on our website and include all related details there? Please also attach:

Full error messages/logs
Screenshots or screen recordings where possible
Steps to reproduce each issue
Your XenForo version and PHP version
Any custom configurations related to the plugin

This will help us investigate the problems properly and provide a faster resolution.
 
Qwatt said:
@Osman,
  1. Scheduled posting issue: For some reason, scheduled postings are not working. They are configured correctly and they do trigger, but the final results are never posted on the forum.

Click to expand...
Thank you! I actually have a guess as to why this isn't working. It looks like we need to allow the same Target forums in both schedules and bots > access > threads.
Osman said:
Hi @Qwatt
Could you please create a support ticket on our website and include all related details there? Please also attach:

Full error messages/logs
Screenshots or screen recordings where possible
Steps to reproduce each issue
Your XenForo version and PHP version
Any custom configurations related to the plugin

This will help us investigate the problems properly and provide a faster resolution.
Click to expand...


Code: 
LogicException: AI Assistant mention handle error: Attempted to set 'status' while a save was pending without forceSet src/XF/Mvc/Entity/Entity.php:645

#0 src/XF/Mvc/Entity/Entity.php(618): XF\Mvc\Entity\Entity->set('status', 'failed')
#1 src/addons/XENTR/AIAssistant/Entity/ActionLog.php(223): XF\Mvc\Entity\Entity->__set('status', 'failed')
#2 src/addons/XENTR/AIAssistant/Service/Content/MentionReplyCreator.php(156): XENTR\AIAssistant\Entity\ActionLog->markAsFailed('MySQL query err...')
#3 src/addons/XENTR/AIAssistant/Job/MentionHandle.php(490): XENTR\AIAssistant\Service\Content\MentionReplyCreator->create()
#4 src/addons/XENTR/AIAssistant/Job/MentionHandle.php(122): XENTR\AIAssistant\Job\MentionHandle->createBotReply(Object(XFMG\XF\Entity\Post), Object(XENTR\AIAssistant\Entity\Bot), 'auto_reply')
#5 src/addons/XENTR/AIAssistant/Job/MentionHandle.php(52): XENTR\AIAssistant\Job\MentionHandle->processPost(5.63269)
#6 src/XF/Job/Manager.php(275): XENTR\AIAssistant\Job\MentionHandle->run(5.63269)
#7 src/XF/Job/Manager.php(205): XF\Job\Manager->runJobInternal(Array, 5.63269)
#8 src/XF/Job/Manager.php(89): XF\Job\Manager->runJobEntry(Array, 5.63269)
#9 job.php(46): XF\Job\Manager->runQueue(false, 8)
#10 {main}


array(4) {
  ["url"] => string(8) "/job.php"
  ["referrer"] => string(40) "https://hhhhhhhhhhhh.hh/threads/1.24953/"
  ["_GET"] => array(0) {
  }
  ["_POST"] => array(0) {
  }
}



LogicException: AI Assistant mention handle error: Attempted to set 'status' while a save was pending without forceSet src/XF/Mvc/Entity/Entity.php:645

#0 src/XF/Mvc/Entity/Entity.php(618): XF\Mvc\Entity\Entity->set('status', 'failed')
#1 src/addons/XENTR/AIAssistant/Entity/ActionLog.php(223): XF\Mvc\Entity\Entity->__set('status', 'failed')
#2 src/addons/XENTR/AIAssistant/Service/Content/MentionReplyCreator.php(156): XENTR\AIAssistant\Entity\ActionLog->markAsFailed('MySQL query err...')
#3 src/addons/XENTR/AIAssistant/Job/MentionHandle.php(490): XENTR\AIAssistant\Service\Content\MentionReplyCreator->create()
#4 src/addons/XENTR/AIAssistant/Job/MentionHandle.php(122): XENTR\AIAssistant\Job\MentionHandle->createBotReply(Object(XFMG\XF\Entity\Post), Object(XENTR\AIAssistant\Entity\Bot), 'auto_reply')
#5 src/addons/XENTR/AIAssistant/Job/MentionHandle.php(52): XENTR\AIAssistant\Job\MentionHandle->processPost(8)
#6 src/XF/Job/Manager.php(275): XENTR\AIAssistant\Job\MentionHandle->run(8)
#7 src/XF/Job/Manager.php(205): XF\Job\Manager->runJobInternal(Array, 8)
#8 src/XF/Job/Manager.php(89): XF\Job\Manager->runJobEntry(Array, 8)
#9 job.php(46): XF\Job\Manager->runQueue(false, 8)
#10 {main}

array(4) {
  ["url"] => string(8) "/job.php"
  ["referrer"] => string(48) "https://hhhhhh.hhh/threads/bhhh.24946/"
  ["_GET"] => array(0) {
  }
  ["_POST"] => array(0) {
  }
}


Your XenForo version and PHP version 8.3.24

Error when generating a reply in a thread containing a guest post:

1779639051631.webp
 
Osman updated [XTR] AI Assistant with a new update entry:

1.0.20

Changelog:​

  • Fixed critical error during AI mention replies — Resolved LogicException: Attempted to set 'status' while a save was pending without forceSet that occurred when the AI mention handler encountered a database or API error. The ActionLog entity now uses forceSet when marking status as failed during error handling, preventing cascading exceptions.
  • Fixed custom tone/prompt not being applied to AI suggestions — The "Suggest Reply" feature...
Click to expand...

Read the rest of this update entry...
 
anurag1 said:
Nice plugin, but don`t you think it`s a bit pricey , i think you should give 50% off for first 20 buyers.
Click to expand...
Thank you! I'm glad you like the plugin.

I understand your point, but the price reflects not only the development time, but also future updates, support, and maintenance. At the moment, I don't have plans for a 50% introductory discount, but I really appreciate the suggestion.
 
Osman updated [XTR] AI Assistant with a new update entry:

1.0.21

We continue to keep pace with the rapid advancements in the AI ecosystem! With version 1.0.21 of XENTR AI Assistant, we are proud to introduce the highly anticipated native Azure OpenAI support for our enterprise users. You can now use your Azure infrastructure directly through our add-on without the need for any intermediary proxies.

Furthermore, this release integrates all the next-generation AI models released as of June 2026 into the core of our add-on. Your assistant now...
Click to expand...

Read the rest of this update entry...
 
Osman updated [XTR] AI Assistant with a new update entry:

1.0.22

In this release, we have taken two massive steps forward in the core architecture of the add-on: We introduced Granular Admin Permissions to ensure complete security and team management, while completely rebuilding the Image Generation infrastructure to be compatible with OpenAI's newest models and third-party proxy services.

✨ Highlighted Features​

1. 🛡️ Comprehensive Admin Panel Permission System Allowing all administrators to modify API keys or...
Click to expand...

Read the rest of this update entry...
 
FrancescoL said:
Also can react to post?
Click to expand...
Currently, the AI Assistant cannot react to posts (e.g., Like, Love). It is designed specifically for content generation such as replying to threads, creating new topics, writing profile posts, and answering support tickets.

The main reason we haven't implemented an automatic reaction feature is API token cost and efficiency. In order for an AI bot to leave an appropriate and context-aware reaction, it would need to actively read and analyze almost every single post made on your forum. For an active community, sending all that text to an AI provider (like OpenAI or Claude) just to determine a reaction would consume a massive amount of API tokens, turning it into a huge and unnecessary expense for forum owners.

By focusing strictly on content generation where the bot is explicitly invoked or set to auto-reply, we ensure that the add-on remains both highly functional and cost-effective for you.
 
Last edited:
Osman updated [XTR] AI Assistant with a new update entry:

1.0.23

In this version, we have rolled out a critical infrastructure update specifically for our Azure OpenAI users, ensuring full compatibility with Azure's next-generation image models.

1. Seamless Azure GPT Image Compatibility We have permanently resolved the "The requested operation is unsupported" error that occurred when trying to generate images using gpt-image-1 or gpt-image-2 deployments on Azure OpenAI. The add-on now automatically...
Click to expand...

Read the rest of this update entry...
 
You must log in or register to reply here.
Back
Top Bottom