VEX Settings | Yaplet

Connections

The AI can be connected to knowledge bases and documentation to provide context-aware responses. These connections are configured in the widget settings.

When you initially set up a knowledge base or documentation for a widget, the AI is automatically trained on all the information contained within it. This training data serves as context to help the AI provide accurate, relevant answers to customer questions. If you later edit or update your knowledge base or documentation content, these changes are automatically synchronized with the AI, ensuring your chatbot always has the most current information without requiring manual retraining.

The connections table shows:

Widget: The connected widget with a link to its settings
Knowledge Base: The knowledge base providing context (if configured)
Documentation: The documentation providing context (if configured)

Each knowledge base and documentation entry includes a "Chunks" button that allows you to preview how the content has been processed and chunked for the AI to understand.

Configuration Options

The VEX settings include several key configuration areas:

Basic Settings

Name: A descriptive name for your chatbot
Active: Enable or disable the chatbot functionality

Context & Branding

Configure company information and communication preferences:

Company name and website for personalized responses
Emoji settings to control AI communication style
Custom prompt for additional instructions

Agent Escalation

Control when and how human agents are involved:

Agent request threshold: Minimum conversation length in the last 5 minutes before offering agent chat
Automatic agent requests: Allow AI to escalate when it determines human help is needed
No context handling: Define behavior when AI lacks relevant information

Privacy & Security

Protect user data and control AI processing behavior:

PII masking: Automatically protect sensitive personal information before sending to AI models
Context reranking: Enable advanced context retrieval for improved response quality

Context Reranking

Context reranking improves the relevance of information retrieved for AI responses by using AI-powered ranking of search results. Costs a little bit more but provides significantly better context matching.

When enabled (default): Provides better context quality but adds 1-1.5 seconds to response time
When disabled: Faster responses with sufficient quality for most use cases

Enable this feature when precision and context quality are critical for your use case.

AI Models

View and configure the AI models used for your chatbot:

Base model: Primary AI model for standard responses
Low model: Alternative model for specific scenarios
Model-specific options are displayed when available

Deleting a Chatbot

Deleting a chatbot permanently removes all its settings and data. This action cannot be undone.

Use the delete button at the bottom of the settings page to remove a chatbot. You'll be asked to confirm the deletion before proceeding.