[XTR] AI Assistant 1.0.11

Fix: Added support for OpenAI Reasoning (o1/o3 series) models. Implemented automatic switching from max_tokens to max_completion_tokens and removal of unsupported creativity parameters (temperature, etc.) for these models.
Fix: Resolved an issue where notifications were not sent to users for replies to Profile Posts.
Improvement: Implemented a safeguard against mass token usage in systems with large Knowledge Bases. The system now prevents loading the entire database for short or empty queries/images, significantly reducing API costs.

This update focuses on compatibility with OpenAI's latest models and significant performance optimizations.

Key Changes:

• OpenAI o1 and o3 Model Support: You can now seamlessly use OpenAI's next-generation "Reasoning" models. The add-on automatically detects these models and optimizes API requests according to the new standards (max_completion_tokens, etc.).
• Smart Token Conservation: A critical improvement for users with large Knowledge Bases (RAG). The bot will no longer attempt to load the entire database when the user message is very short or contains only an image. This effectively prevents unnecessary API costs.
• Notification Fix: Resolved the issue where bot replies to profile posts were not triggering user notifications.

• Fix: Refactored bot triggering and forum permission logic. The "Access > Reply forums" setting is now strictly enforced as a prerequisite for bot triggers (mentions/quotes).
• New Feature: Added "Raw Mode" support to the "Summarize Thread" function. Bots with Raw Mode enabled will now provide direct summaries without conversational fillers.
• Improvement: Reduced the delay timer for the "Reply to unanswered threads" action from 24 hours to 15 minutes, making bots significantly more responsive to new discussions.
• Update: Updated admin panel option phrases to clearly explain the distinction between general access permissions and trigger filters.

This update introduces significant improvements to permission management and bot response times.

• Consistent Permissions: Based on feedback, we've refactored forum access logic. Permissions in the Access tab now act as global rules, while settings in the Activation tab act as additional filters. This resolves configuration conflicts.
• Faster Response: Bots configured to reply to unanswered threads now engage within 15 minutes instead of waiting 24 hours.
• Smarter Summaries: The "Summarize Thread" feature now respects your bot's "Raw Mode" setting, delivering concise, professional summaries without unnecessary chatter.

• Critical: Fixed context limit "0" bug - When "Thread context limit" was set to 0, the system incorrectly sent ALL previous messages to the AI instead of none. This caused translation bots and other specialized bots to ignore their instructions and switch to general chat mode. (<MentionReplyCreator.php, PromptBuilder.php>)
• Critical: Fixed conversation context limit bug - Same issue as above, but for Direct Message conversations. (MentionReplyCreator.php)
• Fixed schedule queue handling - Previously, schedules were marked as "completed" even when no suitable thread was found. Now schedules properly retry on the next run if no eligible content is found. (ScheduleRunner.php)
• Fixed array_rand() empty array error - Server error "array_rand(): Argument #1 ($array) cannot be empty" no longer occurs when node_ids array is empty. (ScheduleRunner.php)
• Fixed 'skipped' status not recognized - The ActionLog entity now properly recognizes "skipped" as a valid status value for skip pattern detection. (ActionLog.php)
• Fixed error_message column reference - Corrected wrong column name that was causing errors when logging skipped responses. (MentionReplyCreator.php)

Important Notes

• Translation bots using "Thread context limit: 0" should now work correctly
• "Reply Unanswered Threads" schedule feature is now more reliable

We're releasing version 1.0.9 with critical bug fixes that address issues reported by our customers.

What's Fixed

1. Translation Bot / Specialized Bot Issues
We discovered and fixed a critical bug affecting bots with strict instructions (like translation-only bots).

The Problem:
When you set "Thread context limit" to 0 (meaning "don't show previous messages to AI"), the system was actually sending all previous messages. This caused the AI to see previous conversations and switch to general chat mode, ignoring your specific instructions.

The Solution:
Setting context limit to 0 now properly means "include zero previous messages". Your specialized bots will now correctly follow their Thread Prompt instructions.

2. Scheduled Tasks Reliability

• Reply Unanswered Threads: Now properly retries if no eligible thread is found, instead of marking the schedule as "completed" and waiting for the next interval.
• Server Errors Fixed: No more "array_rand() empty array" errors in the server log.

3. Skip Pattern Feature
Fixed issues with the [SKIP] response pattern feature that allows AI to skip certain posts conditionally.

Thank You

Thank you to all customers who reported these issues. Your feedback helps us improve the add-on for everyone.

If you encounter any issues after updating, please open a support ticket.

New Features

• Skip Response Feature - AI can now skip responding when conditions aren't met. When AI responds with [SKIP], [NO RESPONSE], [NO TRANSLATION NEEDED], [SILENT], or similar patterns, the bot will silently skip that post. Perfect for translation bots and conditional responders.
• Thread Context Limit Explanation - Added a clear explanation for the "Thread context limit" setting, clarifying that it controls previous messages and the trigger post is always included separately.

Bug Fixes

• Fixed "All" User Group Selection - Completely rewrote the user group access check using XenForo's native pattern (AbstractPrefix). Selecting "All" now works correctly without needing to manually select each group.
• Fixed Empty Trigger Checkboxes - You can now uncheck both "Mention" and "Quote" triggers if you only want auto-reply to work. Previously, the form would reset to default triggers when saved.
• Fixed Language Detection - Bot now responds in the same language as the trigger post (last message), even if previous context messages were in a different language. Added explicit rule to system prompts.
• Fixed Bot Self-Reply in Scheduled Content - Added self-reply prevention to QueueProcessor for scheduled replies. Bot no longer replies to its own previous messages.
• Fixed Bot-to-Bot Loop Prevention - Extended bot loop prevention to scheduled content, respecting the "Prevent bot-to-bot responses" option.

Improvements

• XenForo Standard Compliance - User group access check now uses XenForo's native $user->isMemberOf() method for better compatibility.
• Better Default Values - Default user groups setting is now [-1] (All) matching XenForo's standard pattern.
• Enhanced Thread Prompt Documentation - Thread prompt explanation now includes information about the skip pattern feature.

This update focuses on fixing critical issues reported by users, particularly around automatic translation bot scenarios and user group access.

Translation Bot Support
This version introduces the Skip Response feature, specifically designed for translation bots and conditional responders. When you instruct your AI to respond with [SKIP] for content that shouldn't be translated (e.g., already-English posts), the bot will silently skip without posting a reply.

Critical Bug Fixes
User Group Selection: The "All" option now properly grants access to all users using XenForo's native $user->isMemberOf() method.
Empty Triggers: You can now save a bot with no Mention/Quote triggers enabled if you only want auto-reply functionality.
Language Detection: Bot now correctly responds in the language of the message it's replying to, not the conversation history's dominant language.
Self-Reply Prevention: Bots no longer create consecutive replies in scheduled/auto-reply scenarios.

Translation Bot Setup Guide
For automatic translation bots:

1. Thread context limit: Set to 0
2. Thread Prompt:

Code:

You are a translation bot. ONLY translate non-English posts.
- If English: respond with [SKIP]
- If not English: translate to English
- Do NOT have conversations

3. Triggers: Uncheck both Mention and Quote
4. Auto-reply: Enable and select target forums
5. User groups: Select "All"

Upgrade Instructions
After upgrading, please reconfigure your bots - especially if you were using workarounds for the "All" user group issue.

Thank you to everyone who reported issues!

Bug Fixes

• Fixed "All" User Group Selection - Selecting "All" in the bot's user group access settings now works correctly. Previously, you had to manually select each user group.
• Fixed Bot Self-Reply Issue - Bots no longer reply to their own previous messages in scheduled/auto-reply scenarios. Added proper checks to prevent bot-to-bot loops in scheduled content.
• Fixed AI Self-Quoting - Resolved an issue where AI would mimic the context message format and start responses with "[BotName wrote:]" style prefixes.
• Fixed Language Detection - Bot now responds in the same language as the trigger post (last message), even if previous context messages were in a different language.

Improvements

• Thread Context Limit Explanation - Added a clear explanation for the "Thread context limit" setting. Now clarifies that this controls PREVIOUS messages, and the trigger post is always included separately.
• Suggestion Tones Documentation - Added an explanation to the Suggestion Tones admin page clarifying that these tones are for the "AI Suggest Reply" feature in the editor, not for bot auto-replies.
• Clearer System Prompts - Strengthened AI instructions to prevent unwanted formatting and ensure proper language detection.

This update focuses on fixing reported issues and improving the overall user experience. We've addressed several bugs reported by the community and added better documentation to prevent confusion.

Highlights

Critical Bug Fixes

"All" User Group Selection: Fixed a bug where selecting "All" in the bot's access settings wouldn't work. You no longer need to manually select each user group.

Bot Self-Reply Prevention: Bots will no longer reply to their own previous messages. This was particularly noticeable in scheduled/auto-reply scenarios where the bot might create multiple consecutive replies.

Language Detection: The AI now correctly responds in the language of the message it's replying to, not the dominant language of the conversation history. If someone posts in English, the bot replies in English - regardless of previous messages being in other languages.

We've added explanations to commonly misunderstood settings:

• Thread Context Limit now clearly explains that it controls previous messages, with the trigger post always included
• Suggestion Tones page now explains these are for user-generated content, not bot auto-replies

Upgrade Instructions
We recommend reviewing your bot settings, especially if you were using workarounds for the "All" user group issue.

Thank you to everyone who reported issues and helped improve the add-on!

New Features

• Smart Button Visibility - AI action buttons now automatically hide in forums where no bot is available
• Generate Image, Ask AI, Translate, Summarize, and Reply Suggest buttons respect forum permissions
• Buttons only appear if the bot is configured to operate in that specific forum
• New template function xentr_has_bot() for forum-based visibility checks

Improvements

• Summarize Language Fix - AI now uses user's XenForo language for summaries instead of guessing
• All AI functions now use bot's temperature and max_tokens settings from Intelligence tab
• Removed hardcoded AI parameters for consistent behavior across all features
• Simplified forum permission logic - now uses only "Reply forums" field from Access tab

Technical Changes

• Added templater_setup event listener for custom template functions
• Created Template/TemplaterSetup.php class for template helpers
• Added hasAvailableBotForNode() method to BotRepository
• Updated canOperateInNode() to use simplified logic

Bug Fixes

• Fixed Summarize output being truncated due to hardcoded max_tokens value
• Fixed ProfilePostCreator using incorrect property access for bot settings
• Fixed inconsistent temperature values across different AI features

This release focuses on improving user experience with smarter button visibility and fixing the language detection for AI summaries.

Highlights

1. Smart Button Visibility
AI action buttons (Generate Image, Ask AI, Translate, Summarize, Reply Suggest) now intelligently hide in forums where no bot is configured to operate.

How to Configure:

1. Go to Admin Panel → AI Assistant → Bots → Edit
2. Navigate to the Access tab
3. Select forums in the Reply forums field
4. Save

If "Reply forums" is empty, the bot works everywhere. If specific forums are selected, buttons only appear in those forums.

2. Summarize Language Fix
The AI summary feature now correctly uses the user's selected XenForo language instead of sometimes responding in English.

3. Consistent AI Parameters
All AI features now properly use the bot's temperature and max_tokens settings from the Intelligence tab. Previously, some features used hardcoded values.

Upgrade Notes

• Review your bot's "Reply forums" setting if you want forum-specific button visibility

New Features

• Customizable Suggestion Tones - New admin panel section for managing AI response styles
• Create unlimited custom tones with personalized prompt instructions
• Enable/disable tones with a single click toggle
• Set any tone as the default selection
• Pre-installed default tones: Friendly, Professional, Humorous, Sarcastic

Improvements

• AI suggestion system now uses dynamic tone instructions from database
• Added tone information display in Activity Log details view
• Increased max_tokens to 1500 for longer AI responses
• Improved tone parameter handling between editor and controller

Database Changes

• New table: xf_xentr_aia_tone
• New registry cache: xentrAiaToneCache

This release introduces Customizable Suggestion Tones, allowing administrators to fully customize AI response styles directly from the admin panel.

What's New?

You can now create your own AI tones from the admin panel. Define custom prompt instructions for each tone, enabling your users to receive AI suggestions in different styles.

For example:

• Agree Message - Responses that positively contribute to the discussion
• Disagree Message - Responses that politely present a different perspective
• Asking Message - Responses that ask curious questions

How to Use

Navigate to Admin Panel > AI Assistant > Suggestion Tones and create your custom tones. Users can select their preferred tone from the dropdown menu below the editor to receive tailored AI suggestions.

Upgrade Notes

• Existing installations will automatically receive default tones via upgrade
• Database changes are applied automatically
• Backward compatibility is maintained

Bug Fixes

• UTF-8 Compatibility - Fixed "Incorrect string value" error during installation caused by emoji characters on some hosting environments.
• Profile Post Notifications - Fixed notification system for scheduled profile posts. Users will now receive alerts when a bot posts on their profile.
• Activity Log Links - Content link was missing for "Create profile post" actions. You can now click the link icon to go directly to the profile post.
• Action Type Label - Added missing label for "create_profile_post" in Activity Log.

Improvements

• Schedule List - Behavior Profile name is now displayed in the schedule list. This makes it easier to identify what each schedule does.

This release addresses installation issues, reported problems with the profile post feature, and improves schedule management.

UTF-8 Compatibility Fix
Some hosting environments were experiencing "Incorrect string value" errors during installation. This issue was caused by emoji characters requiring UTF-8 4-byte (utf8mb4) database support. The add-on is now compatible with all MySQL configurations.

Profile Post Notifications
Scheduled profile posts created by bots now trigger user notifications. Previously, these posts were created silently. Now they trigger notifications just like regular profile posts.

Activity Log Improvements
Added missing content link for "Create profile post" actions in Activity Log. You can now click the log entry to go directly to the related profile post.

Schedule Management
Behavior Profile name is now displayed in the schedule list:

Code:

XenBot → Create profile post → Welcome Greeter

This makes it easier to identify what each schedule does when you have multiple schedules.

New Features

• Profile Post Creation - Scheduled tasks can now automatically post messages to user profiles
• User Criteria System - Target user selection for profile posts (specific users, user groups, registration date, etc.)
• New action type: "Create profile post" scheduling option
• Custom icon and label for profile posts in Activity Log

Improvements

• Thread tools menu "Enable/Disable AI Bots" toggle now updates dynamically
• Added "Profile prompt" field to Behavior Profiles

Bug Fixes

• Fixed AI provider "generate()" method call error
• Fixed user criteria saving and reading issues
• Fixed thread toggle switch text update problem

This release introduces a new scheduling feature that allows your AI bots to automatically post messages on user profiles.

Profile Post Creation
Your bots can now post welcome messages, birthday greetings, or motivational messages to user profiles based on your defined criteria.

Example Use Cases:

• Automatic welcome messages for new members
• Birthday greetings for users
• "We miss you" messages for inactive users
• Thank you and congratulation messages for active members

How to Use

1. Create a new profile in Behavior Profiles and fill in the "Profile prompt" field
2. Add a new schedule in Schedules
3. Select "Create profile post" as the action type
4. Define target users with User Criteria
5. Save and wait for the schedule to run

• Moved the Personality field to the Intelligence tab in the bot edit page for better workflow.
• Refactored the Personality and Context-Specific Prompts (Thread, DM, Profile) logic. Personality and context instructions are now correctly merged in bot responses.
• Added a new option for selecting a Utility Bot. You can now designate a specific bot for tasks like title suggestions, summarization, and translation, or leave it on "Auto" to use the first available bot.
• Implemented a user-friendly dropdown menu for the Utility Bot selection option.
• Fixed the context format ([User wrote:]) used by AI models when analyzing past messages. This resolves the issue where AI would mimic this format and "self-quote" in its responses.
• Strengthened the System Prompt structure with new rules preventing the AI from starting responses with prefixes or usernames.
• Updated and improved various phrases.

This update brings critical fixes and improvements to bot personality and context management. We focused specifically on the functionality of the "Personality" field and the consistency of bot conversation flows.

Highlights:

Enhanced Personality & Context Management
Bot personality now merges seamlessly with context-specific instructions (Thread, DM, Profile Prompts). The Personality field has been moved to the "Intelligence" tab in the bot edit screen, where it logically belongs.

Customizable Utility Bot
For background tasks (title suggestions, summarization, etc.), you can now assign a specific bot (e.g., a "System Bot") instead of using a random one. A new dropdown menu in the settings page makes this selection effortless.

"Self-Quoting" Issue Fixed
We resolved a formatting issue where bots would sometimes start responses by quoting themselves (e.g., "Death wrote:"). The context format and system instructions have been updated to ensure bots provide much cleaner and more natural responses.

Update Instructions:
Please check your bot settings after updating. We recommend reviewing the new Personality field location under the "Intelligence" tab.