HELP CENTER

Plugin Doc Q&A

Create AI documentation assistants that answer questions about your manuals, guides, and documents.

What is Doc Q&A?

Doc Q&A allows you to create specialized AI assistants that answer questions based on your documents. Each assistant can have its own set of files and a customized prompt.

  • Create multiple assistants for different purposes
  • Upload PDF, DOCX, TXT, MD and more documents
  • Generate an embeddable widget for your website
  • Check the history of questions and answers
  • Ideal for help centers, technical support and FAQs

Use cases

📚

Technical documentation

User manuals, installation guides, technical specifications. Users can ask how to configure X and get precise answers.

Help Center

Frequently asked questions, policies, terms and conditions. Reduce support tickets with automated responses 24/7.

🎓

Training and courses

Educational materials, study guides. Students can ask questions about the course content.

📋

Internal processes

Procedures, regulations, employee guides. Your team can quickly look up processes.

How to create an assistant

  1. 1

    Access the plugin

    On the Dashboard, go to Plugins → Doc Q&A

  2. 2

    Create a new assistant

    Click on “New assistant“ and give it a descriptive name (e.g.: “Technical Support“, “Business FAQs“)

  3. 3

    Set up the prompt

    Write the instructions for the AI: tone, restrictions, how it should respond. Define its personality and limits.

  4. 4

    Upload your documents

    Drag or select the files that contain the information. Formats: PDF, DOCX, TXT, MD, CSV, JSON, HTML.

  5. 5

    Activate the assistant

    Once the documents have been processed (this may take a few seconds), activate the assistant so it can start working.

  6. 6

    Get the embed code

    Copy the integration code and paste it into your website where you want the assistant to appear.

Integration code

Copy the assistant’s embed code and paste it wherever you want it to appear:

<!-- WAzion Doc Q&A -->
<div id="wazion-doc-qa"></div>
<script src="https://www.wazion.com/plugins/doc-qa/embed.js"
        data-assistant="TU_ASSISTANT_ID"></script>

The assistant’s ID is generated automatically when it is created. You will find it in the assistant’s settings.

Widget dimensions

The widget adapts to the container where you insert it. You can control its size with CSS:

<div id="wazion-doc-qa" style="height: 500px; max-width: 600px;"></div>

Document management

Supported formats

PDF DOCX TXT MD CSV JSON HTML XML

Limits

  • Maximum 50 MB per file
  • Up to 1GB of total storage per store (shared with Knowledge Base)
  • No limit on files per assistant

Processing states

Processing The file is being analyzed and vectorized
Ready The file is available for consultation.
Error The file could not be processed.

Tip: Descriptive titles

When uploading a file, give it a descriptive title. The AI uses the title to decide which document to search. “Manual de instalación v2.pdf“ is better than “documento1.pdf“.

Question history

Each assistant stores a history of all questions received. From the Dashboard you can:

  • View the latest questions with their answers
  • Check consumed tokens and response time
  • Identify frequently asked questions that you could add to the documents.
  • Detect incorrect answers to improve the documents

Statistics

On each assistant’s panel you will see key metrics:

?
Total questions
?
Last 7 days
?
Average time
?
Documents

How does it differ from other products?

Feature Doc Q&A Knowledge Base Product Q&A
Purpose Independent assistants Enrich WhatsApp/Chat Q&A for Shopify products
Data source Uploaded documents Uploaded documents Shopify data
Multiple instances
Embeddable widget
Requires Shopify No No Yes

Frequently Asked Questions

Related articles

WAzion Assistant

Commercial information and technical support

Hello! I am the WAzion assistant. I can help you with information about prices and plans, technical questions, configuration, or any questions about our product. How can I assist you?
Developed with WAzion AI