Contact Us

If you still have questions or prefer to get help directly from an agent, please submit a request.
We’ll get back to you as soon as possible.

Please fill out the contact form below and we will reply as soon as possible.

  • Contact Us
English (US)
US English (US)
CO Spanish (Colombia)
  • Home
  • wvx CRM

How to configure and use the wvx CRM Assistant (Knowledge Retrieval) with artificial intelligence in wolkvox CRM

Written by Jhon Bairon Figueroa

Updated at November 13th, 2025

Contact Us

If you still have questions or prefer to get help directly from an agent, please submit a request.
We’ll get back to you as soon as possible.

Please fill out the contact form below and we will reply as soon as possible.

  • wvx Digital Interaction
  • wvx Voice Interaction
  • wvx Conversational AI
  • wvx Control Interactions
  • wvx CRM
  • wvx Agent
  • wvx Studio
  • Cibersecurity and Compliance
  • Release
    wolkvox Manager release wolkvox Agent release wolkvox CRM release
+ More

Table of Contents

Introduction Configuration

Introduction

wolkvox CRM allows you to integrate artificial intelligence to assist users with relevant business information, automate responses, and improve operational efficiency. I want to create, configure, and manage specific knowledge bases (RAGs) for different teams or areas, and activate the AI assistant so users can consult it in real-time.

"Knowledge Retrieval" and "wvx CRM Assistant" are integrated tools in wolkvox CRM designed to enhance user assistance through artificial intelligence. They allow you to create "RAGs" (Retrieval Augmented Generation), which are knowledge bases trained with specific business information. This segmentation is crucial so that each team (sales, collections, customer service, technical support, etc.) has access to accurate and relevant information for their functions.

The "wvx CRM Assistant" is the chat interface where users interact with the AI, obtaining responses based on the configured knowledge base. This assistant is designed to provide quick and specific support in various situations within the wolkvox CRM modules, without maintaining a prolonged conversation history to focus on specific inquiries.

     

Use cases:

  • A sales agent needs quick information about a specific product or discount policy.    
  • A support agent requires technical data to troubleshoot a common problem.    
  • A collections agent needs to verify details about a payment plan or a portfolio case.    
  • Automate responses to frequently asked questions to free up staff time.    

     

 

Configuration

To configure and use the wvx CRM Assistant, follow these steps:

     

Accessing Knowledge Retrieval Configuration:

  1. In the left sidebar menu of wolkvox CRM, click the "Configuration" button.    
  2. Within the Configuration section, select the "Knowledge Retrieval" tab.    

     

 

Creating a new RAG (Knowledge Base):

In the "Knowledge Retrieval" interface, in the bottom right, click the " + Create rag" button.

     

 

  1. A modal window will open. Complete the following fields:        
    1. RAG Name: Assign a unique name to the RAG (only letters, numbers, and underscore; no spaces or special characters).            
    2. Description: Provide a brief description of the purpose or content of the RAG.            
    3. Tags: Type relevant tags and press *Enter* after each one to assign them. This helps categorize and organize your RAGs.            
  2. Click the "Create" button.    
  3. The new RAG will appear in the "RAG list" table.    

     

 

Configuring an existing RAG:

  1. In the "RAG list" table, left-click on the "RAG Name" you want to configure. This will take you to the configuration screen for that RAG.    
  2. Alternatively, you can click the three-vertical-dot button to the right of a RAG in the list and select "Configuration."    

     

 

"RAG Configuration" Tab:

Once inside the RAG configuration, in the "RAG Configuration" tab, adjust the following parameters:

  1. AI Model:        
    1. Select "Fast": For quick and direct responses.            
    2. Select "Thinking": Ideal for situations that require a more detailed context analysis and reasoning.            
  2. Instructions: Write clear instructions for the artificial intelligence. For example: “Provide detailed information based solely on the entered documentation. Do not invent information. If you do not find information related to the request, reply that you have no data regarding it.”    
  3. Max no. results: Set the maximum number of results the AI should consider. The value must be between 5 and 100. (Recommendation: Test different values to find the optimum).        
    1. Error message if the range is incorrect: “The topk must be a number between 5 and 100.”            
  4. Accuracy level (%): Define the desired accuracy level for the responses, which can optimize costs or improve accuracy. The value must be between 0 and 2 (with decimals). (Recommendation: Conduct tests to identify the ideal level).        
    1. Error message if the range is incorrect: “It must be a number between 0 and 2 with one decimal place.”            
  5. Click the "Save" button to apply the changes.    

     

 

Uploading documents to a RAG ("RAG Configuration" or "Knowledge management" Tab):

You can upload documents from two locations:

  1. From the "RAG Configuration" tab: Click the "Upload documents" button (below the "Instructions" field).    
  2. From the "Knowledge management" tab: In the upper right corner, click the "+ Upload document" button.    

     

 

In both cases, a modal window will open:

  • Categories: Enter relevant categories for the document. Press *Enter* after each category.    
  • Chunk quantity: Define the size of the text "chunks" (fragments). A common and recommended value is 500 chunks, but you can set a value between 100 and 4096. (Recommendation: Test and adjust according to the document type and desired responses).    

Select or drop documents to upload: Drag and drop the file or click to select it.

  • Supported formats: PDF, TXT.    
  • Maximum size per file: 50 MB.    
  • Note: Only one document can be uploaded at a time.    

     

 

Managing documents in the "Knowledge management" tab:

  1. In the RAG's "Knowledge management" tab, you will see a table with all the uploaded documents. The columns are: 'Title,' 'Category,' 'Upload date,' and 'Size.'    

For each document, there is a three-vertical-dot button on the right side with the following options:

  • Details: Shows a modal window with basic file information.    
  • Edit: Allows editing only the categories assigned to the document. You cannot change the title, file name, or replace the document.    
  • Delete: Allows deleting the document from the RAG.    

   

  

Testing the configuration:

Use the test chat message thread in the "RAG Configuration" tab to validate the RAG's behavior before activating it for end-users.

     

 

Activating the wvx CRM Assistant for users:

Once you have configured your RAG and uploaded the documents, you must activate the assistant for users:

  1. Return to the left sidebar menu and click "Configuration."    
  2. Go to the "General" tab.    
  3. Within the "General" tab, select "CRM Assistant."    
  4. In the "Select knowledge retrieval base" field, choose the RAG you want to use for the AI assistant.    
  5. Activate the service by clicking the "Activate wvx CRM Assistant" toggle button.    

     

 

Using the wvx CRM Assistant:

  1. Once activated, wolkvox CRM users will see a button in the upper right corner of the interface called "wvx CRM Assistant."        
    • Click this button to open the assistant's chat window.            
  2. Type your message or question in the "Type your message..." field and press *Enter* or click the send icon.        
    • The assistant will respond based on the information contained in the configured RAG.            
    • Each response from the wvx CRM Assistant includes a button to copy the message.            
    • Note that the chat does not maintain a persistent history; it resets with each page reload, as it is designed for specific, one-time assistance.            

     

     

crm knowledge ai assistant

Was this article helpful?

Yes
No
Give feedback about this article

Related Articles

  • How to configure dynamic notifications with attachments in wolkvox CRM
  • How to create and manage user profiles in wolkvox CRM

2025 Wolkvox

Information security policy | Privacy Policy

Expand