VIP Addons XTR - AI Assistant 1.0.29

With this release, we have introduced massive structural improvements to our AI Assistant add-on, specifically focusing on the **Image Generation** capabilities! You can now seamlessly integrate and experience the newest generation of multimodal models offered via **OpenRouter** (including the GPT-5 Image series, Google Gemini Image suite, and Meta Muse Spark)!

Furthermore, trigger flexibility has been dramatically improved! We completely eradicated the frustrating issue where bots would accidentally reply with conversational text descriptions instead of generating actual drawings when prompted. Bots will now instantly identify your drawing commands anywhere inside a prompt and deliver stunning artwork!

[NEW] OpenRouter Multimodal Image Engine (message.images Array Parser):
Engineered an advanced parsing module capable of detecting and extracting generated graphics from the modern message.images data arrays returned by state-of-the-art OpenRouter chat/completions endpoints (such as GPT-5 Image and Gemini Image). Seamlessly supports both Base64 Data URIs and remote HTTP image urls!
[NEW] Added Latest Supported OpenRouter Image Models:
Revamped the image model selection dropdown in the Admin Control Panel (Bot -> Capabilities tab) to include highly economical and lightning-fast modern vision models:
OpenAI: GPT-5 Image Mini, GPT-5 Image, GPT-5.4 Image 2
Google: Gemini 2.5 Flash Image, Gemini 3.1 Flash Image, Gemini 3 Pro Image
Meta: Muse Spark 1.1
[ENHANCEMENT] Smart & Flexible Trigger Pattern Detection:
Removed the strict requirement (str_starts_with) where image triggering keywords (e.g., "create image:", "draw:") had to be positioned at the exact beginning of a sentence. Powered by our new substring scanner (mb_strpos), even if users prepend mentions, greetings, or extra words before the command, the bot instantly intercepts the trigger and routes it to the image creation engine.
[FIX] Resolved Bots Responding with Text Instead of Drawing Images:
Fixed an architectural logic flaw where uncaptured trigger prompts would fall through to standard conversational completion APIs, causing bots to describe imaginary photos with plain text instead of drawing them.
[FIX] Critical ImageFactoryService Bug & Silent Failure Handling:
Resolved a critical backend bug where generating images via background mentions caused a silent PHP fatal error (Class Not Found) due to an outdated class namespace reference (ImageFactory). Upgraded overall exception handling to capture low-level PHP \Throwable events cleanly without breaking queue execution.

Source: xentr.net

• Fix: Added a specialized filtering method to automatically remove Perplexity/Sonar style footnote markers from AI output.
• Fix: Implemented a sanitization step to purge unclosed or broken BBCode tags generated by AI models.
• Fix: Improved the quoting logic to completely remove media tags from the shortened quote content, preventing broken embeds in the front end.
• Enhancement: Added an "IDENTITY REMINDER" reinforcement block to the system prompt to prevent context loss for models.
• Feature: Added a new global option to manage the default checked/unchecked state of the AI trigger on the thread creation page.
• Update: Updated the thread creation template modifications to reflect the new dynamic checkbox option.

With this update, we have focused on improving the robustness of AI outputs, fixing quote rendering issues, and expanding customization options for administrators based on your feedback.
Key Highlights in this Release:

• Smart Citation Filtering: We have added a smart filter for research-augmented models like Perplexity and Sonar. Ugly, unclickable reference markers (like [1] or [1, 2]) that these models sometimes inject are now automatically stripped before rendering on the forum.
• Media Rendering in Quotes: Fixed a critical issue where the AI would sometimes break MEDIA, IMG, or ATTACH BBCode tags when trying to quote a user. Orphaned or half-written tags generated by the AI are now properly cleaned up and sanitized.
• Enhanced AI Persona Retention: Some AI models with shorter context windows tend to "forget" their assigned persona (e.g., "You are a friendly support assistant") as the conversation gets longer. We've added a strict reinforcement block to the end of the system prompt to guarantee your bot stays in character at all times.
• New Admin Option for Thread Creation: You can now control the default behavior of the "Enable AI Bots" checkbox when users create new threads. By using the new "Enable AI Bots by default" option in the admin panel, you can choose to have it unchecked natively.