Hello vBulletin Community!

I’m excited to introduce SafePost AI Filter, a new plugin designed to help forum administrators automatically detect and manage harmful or toxic content on your forums using the Perspective API which is provided for free by Google.

Key Features:

• AI-Powered Content Filtering: Detects offensive, insulting, or toxic content using AI and Machine Learning before it gets published. Spam posts are excluded.
• Flexible Filter Settings: Configure filters individually (Toxicity, Severe Toxicity, Identity Attack, Profanity, Threat, and Insult)..
• Adjustable Threshold: Each filter has a threshold setting (percentage-based), giving you full control over how strict the moderation should be.
• Customizable Actions: Decide whether filtered posts should be blocked with an error or automatically sent for moderation.
• Referennce Codes for Transparency: When a post is blocked, a reference code (e.g., IA0092) shows which filter was triggered and the likelihood score, so you can investigate or assist users quickly.
• Multi-language Support: ​See Language Support section below.
• User-Based Filtering: Apply filters based on usergroups, post count, or both, giving you granular control over who gets moderated.

Why Use SafePost AI Filter Plugin?

• Protect your community: Automatically catch problematic content before it spreads.
• Save moderator time: Reduce manual moderation workload.
• Improve member experience: Prevent toxic content from disrupting your forum.
• Customizable for your forum: Flexible settings allow you to adapt the plugin to your forum’s unique needs.

​
Language Support:
Supports multiple languages, ensuring accurate detection even on multilingual forums. These are the languages supported by the Perspective API.

• Arabic (ar)
• Chinese (zh)
• Czech (cs)
• Dutch (nl)
• English (en)
• French (fr)
• German (de)
• Hindi (hi)
• Indonesian (id)
• Italian (it)
• Japanese (ja)
• Korean (ko)
• Polish (pl)
• Portuguese (pt)
• Russian (ru)
• Spanish (es)

Text Analyzer Tool: (new in v1.1)
The latest update v1.1 adds a Text Analyzer tool in the Admin Control Panel — a quick and easy way for Administrators to test the Perspective API directly within vBulletin AdminCP.

Simply enter any sample text and instantly view the API’s analysis scores for each attribute (Toxicity, Severe Toxicity, Identity Attack, Insult, Profanity and Thread).
​​
This helps you fine-tune your filter thresholds without needing to create a test post in the forum.

Perspective API :
Perspective is a free API that helps you host better conversations online. The API uses machine learning models to score the perceived impact a comment might have on a conversation. These are the prerequisites In order to use this API:

1. A Google account.
2. A Google Cloud project to authenticate your API requests.
3. API access requested from Google.

Follow these detailed instructions to get started. This is also linked from the plugin settings.​

Compatibility:

• Compatible with vBulletin 5 and 6 (self-hosted)
• Not compatible with vBCloud.
• Make sure your forum language(s) are listed in the Language Support section above.

Version History:
v1.1

• Added Text Analyzer tool in AdminCP to enable Admins to quickly test the Perspective API.
• Set "doNotStore" flag to true in the Perspective API request to indicate that the post content should not be stored and used for future research to improve the API.
• Stripped out unnecessary HTML tags in a post before sending it to the Perspective API for toxicity analysis.

v1.0

• Initial version.

Demo:


Price:
$50 (Branding-free)​


​
SafePost AI Filter is ideal for forums of any size and provides a modern, AI-driven approach to keeping your community safe and welcoming.

Ready to get started? Purchase SafePost AI Filter today and give your forum moderation a powerful AI boost!