[XTR] AI Assistant [Paid] 1.0.35

[Qwatt]

Yes, the add-on fully supports this use case! You can create two distinct bot profiles linked to your @Global and @Neighbor XenForo user accounts and assign your custom philosophy prompts to each one individually. When a forum member mentions both bots in a single post, both bots will independently trigger and post their opposing arguments while taking the thread context into account. Please note that to prevent infinite reply loops and unexpected API token costs, the add-on includes a built-in "Prevent bot-to-bot responses" option enabled by default, which you can configure according to how freely you want them to debate each other.

Hi, @Osman, thanks for the great work!
Could you please let me know if it's possible to specify separate SOCKS5 proxies for certain bots or models?
If this cannot be done via the admin panel, could you point out where in the template code and in which specific template this can be implemented? Also, if possible, I would really appreciate it if you could provide an example of such a modification. Thank you so much in advance!

---

 

[JBS]

Just purchased this and I am already very impressed. Definitely just experimenting so far... but very cool.

 

[Osman]

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

1.0.32

This release elevates your administrative experience with interactive analytics, high-definition charts, and sleek metric cards while refining the core content generation engine for unmatched reliability.

Key Highlights:

• Next-Gen Administrative Dashboard:
  The administrative interface has received a complete overhaul. Enjoy dynamic time-series activity trend charts with multi-period filters (24 Hours, 7 Days, 30 Days, All Time), high-DPI distribution donuts, executive...

Read the rest of this update entry...

 

[Osman]

Hi, @Osman, thanks for the great work!
Could you please let me know if it's possible to specify separate SOCKS5 proxies for certain bots or models?
If this cannot be done via the admin panel, could you point out where in the template code and in which specific template this can be implemented? Also, if possible, I would really appreciate it if you could provide an example of such a modification. Thank you so much in advance!

---

Hi @Qwatt, thank you for the kind words!

First, just to clarify a quick technical detail: XenForo templates only render the front-end HTML/CSS layout in the browser. Outgoing AI requests (to OpenAI, Claude, Gemini, etc.) are executed on the server side via PHP using XenForo's underlying Guzzle HTTP client. Therefore, proxy routing cannot be configured inside XenForo template modifications.

Currently, there isn't a dedicated per-bot/per-provider SOCKS5 proxy field in the Admin Control Panel. However, you can easily achieve this using one of the following two methods:

Method 1: Global XenForo Proxy (No add-on modification needed)​

If you want all outgoing XenForo HTTP traffic to route through your SOCKS5 proxy, you can simply add this line to your src/config.php:

$config['http']['proxy'] = 'socks5h://username:password@proxy-ip:port';

Method 2: AI Provider-Specific Proxy (Quick PHP Modification)​

If you only want the AI Assistant requests routed through the SOCKS5 proxy, you can modify the base provider service:

File: src/addons/XENTR/AIAssistant/Service/AI/AbstractProvider.php
Method: sendRequest()

Around line 74, locate $client->post(...) and add the 'proxy' option to the request array:

$response = $client->post($url, [
    'json' => $data,
    'headers' => $mergedHeaders,
    'timeout' => $timeout,
    'connect_timeout' => min(30, $timeout),
    'http_errors' => false,
    'proxy' => 'socks5h://username:password@proxy-ip:port', // <-- Your SOCKS5 proxy URL
]);

If you want to restrict the proxy to a specific provider (e.g., only OpenAI):

$requestOptions = [
    'json' => $data,
    'headers' => $mergedHeaders,
    'timeout' => $timeout,
    'connect_timeout' => min(30, $timeout),
    'http_errors' => false,
];

if ($this->provider && $this->provider->provider_id === 'openai') {
    $requestOptions['proxy'] = 'socks5h://username:password@proxy-ip:port';
}

$response = $client->post($url, $requestOptions);

Roadmap Note:
This is a great feature suggestion! We have added an optional "HTTP / SOCKS5 Proxy" configuration field directly into the AI Provider settings in ACP to our roadmap for upcoming updates so that you won't need manual code edits in the future.

Let me know if you have any questions or need further guidance!

 

[Osman]

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

1.0.33

With this milestone update, our primary focus has been completely revamping and perfecting the AI Discussion Summary system, making it an indispensable, natural, and lightning-fast part of the forum discussion experience.

---

What's New & Highlighted Features​

1. Native & Modern XenForo Interface
The summary box has been rebuilt from the ground up using XenForo's native core styling. Say goodbye to clunky, heavy custom interfaces! The new card design...

Read the rest of this update entry...

 

[Qwatt]

Hi @Qwatt, thank you for the kind words!

First, just to clarify a quick technical detail: XenForo templates only render the front-end HTML/CSS layout in the browser. Outgoing AI requests (to OpenAI, Claude, Gemini, etc.) are executed on the server side via PHP using XenForo's underlying Guzzle HTTP client. Therefore, proxy routing cannot be configured inside XenForo template modifications.

Currently, there isn't a dedicated per-bot/per-provider SOCKS5 proxy field in the Admin Control Panel. However, you can easily achieve this using one of the following two methods:

Method 1: Global XenForo Proxy (No add-on modification needed)​

If you want all outgoing XenForo HTTP traffic to route through your SOCKS5 proxy, you can simply add this line to your src/config.php:

$config['http']['proxy'] = 'socks5h://username:password@proxy-ip:port';

Method 2: AI Provider-Specific Proxy (Quick PHP Modification)​

If you only want the AI Assistant requests routed through the SOCKS5 proxy, you can modify the base provider service:

File: src/addons/XENTR/AIAssistant/Service/AI/AbstractProvider.php
Method: sendRequest()

Around line 74, locate $client->post(...) and add the 'proxy' option to the request array:

$response = $client->post($url, [
    'json' => $data,
    'headers' => $mergedHeaders,
    'timeout' => $timeout,
    'connect_timeout' => min(30, $timeout),
    'http_errors' => false,
    'proxy' => 'socks5h://username:password@proxy-ip:port', // <-- Your SOCKS5 proxy URL
]);

If you want to restrict the proxy to a specific provider (e.g., only OpenAI):

$requestOptions = [
    'json' => $data,
    'headers' => $mergedHeaders,
    'timeout' => $timeout,
    'connect_timeout' => min(30, $timeout),
    'http_errors' => false,
];

if ($this->provider && $this->provider->provider_id === 'openai') {
    $requestOptions['proxy'] = 'socks5h://username:password@proxy-ip:port';
}

$response = $client->post($url, $requestOptions);

Roadmap Note:
This is a great feature suggestion! We have added an optional "HTTP / SOCKS5 Proxy" configuration field directly into the AI Provider settings in ACP to our roadmap for upcoming updates so that you won't need manual code edits in the future.

Let me know if you have any questions or need further guidance!

Thank you, your plugin is a very powerful tool.
And it is especially unfortunate if a certain provider happens to be unavailable from the hosting server's IP address. Giving up on your plugin shouldn't be the solution. In this case, a proxy will help resolve all such issues.
It is great news that this features will be added to the admin panel in the future.


Method 2 should work for me.

 

[Osman]

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

1.0.34

With this update, we are introducing an intelligent external file search capability that provides significant efficiency and cost savings, particularly for forums that manage extensive documentation, rulebooks, or lengthy manuals.

Previously, when large documents were assigned to a bot, the entire text was loaded and processed on every single reply. For communities with comprehensive knowledge bases, this could lead to excessive token consumption and high operational costs. With version...

Read the rest of this update entry...

 

[expanserpb]

Quick question: if I have a large knowledge base in a XF wiki addon, what would it take to enable the assistant to be able to retrieve from this?

 

[Osman]

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

1.0.35

This release introduces essential bug fixes and improvements covering vector knowledge base responses, the bot selection menu, bot reply reliability when mentioned, and AI image generation model handling.

An issue where vector knowledge base (Vector Store / RAG) responses occasionally included unwanted citation tokens, file reference tags, and source brackets has been resolved. Bot answers are now automatically sanitized, delivering clean and polished responses without leaking internal...

Read the rest of this update entry...