AI Product Support for Magento 2
YOU CAN TRUST US
25 years of experience in e-commerce and Magento 2
Fast delivery
Efficient implementation process
Simple and transparent complaint process
Working with clients worldwide
Free module updates
Payment by bank transfer
What this module is
AI Product Support is a Magento 2 module that adds a convenient AI chat to your store.
It can work in two places:
- in the Magento Admin Panel,
- optionally on the storefront for customers.
Its purpose is to quickly answer questions about products, modules, documentation, and help content stored in an AI knowledge base prepared for the store.
Important distinction: this module does not organize store content from scratch. It exposes already prepared knowledge through a convenient chat interface.
What it delivers in practice
In daily work, the module reduces the time spent searching for information. Instead of browsing descriptions, manuals, FAQ, and notes across different places, the user can simply ask in the chat.
Example use cases:
- a team member asks about product features,
- support checks what a given module includes,
- sales looks for a quick answer for a customer,
- a customer asks about a product, use case, or basic information on the storefront.
For which stores
The module is designed to work across different types of Magento 2 stores, for example:
- stores selling modules and extensions,
- stores with a broad product catalog,
- stores with their own documentation and FAQ,
- stores that want to add a simple AI support layer without rebuilding the entire site.
This is not a solution for only one specific project. It is a module that can be adapted to many stores as long as they have a prepared AI knowledge base.
How this module connects to the store knowledge base
For the chat to work well, the store first needs structured content.
In practice this includes materials such as:
- product descriptions,
- attributes,
- FAQ,
- manuals,
- documentation,
- additional help content.
The Kowal AI Product Feed module page describes it accurately: its role is to convert store product data into a structured, up to date, and AI ready knowledge base. This step creates the foundation for later AI features in the store. Source: Kowal AI Product Feed for OpenAI Vector Store
The simplest way to understand it:
- one module organizes and prepares knowledge,
AI Product Supportuses that knowledge in a conversation with the user.
Thanks to this, the chat is not detached from the store. It uses materials prepared specifically for the given catalog and content set.
Key features
1. AI chat in the Admin Panel
In Magento Admin, a permanent AI tab appears on the right edge of the screen. It is visible across multiple admin sections and opens a slide out chat panel.
This provides AI assistance from almost anywhere in the backend:
- while managing products,
- while processing orders,
- while working on content,
- while reviewing documentation,
- while supporting a customer.
2. Optional customer chat on the storefront
The module can also be enabled on the storefront. Then customers see a simple AI tab and can ask a question without leaving the page.
This works especially well where customers often ask about:
- differences between products,
- core features,
- compatibility,
- documentation,
- how the product works.
3. Answers grounded in the store knowledge base
The module does not behave like a generic chatbot. It is designed to answer primarily based on knowledge provided by the store.
That means:
- better alignment with the store offer,
- lower risk of guessing,
- better fit to real products and content,
- more control over what the AI shows to the user.
This is the biggest difference between a typical AI chat and an AI solution sensibly implemented in an online store. First you prepare the knowledge, and only then you expose it to users through chat.
4. Readable response panel
Answers are displayed in a convenient format:
- split into paragraphs,
- with bullet lists,
- with emphasis on the most important information,
- with links when needed.
If the question concerns a specific product, the module can additionally display a product cards section, for example:
- image,
- name,
- SKU,
- product link.
5. Dedicated module settings
The module has its own configuration, so it can be managed independently from other extensions.
The administrator can configure, among other things:
- access to the AI service,
- response model,
- knowledge source used by the module,
- request limits,
- message length,
- storefront enablement,
- basic security rules.
6. Abuse protection
If the chat runs on the storefront, the module helps reduce unnecessary load and accidental abuse.
It includes, among other things:
- request count limits,
- control of the first submission after entering the page,
- additional protection against simple bots,
- message length control.
This helps control costs and stability.
Benefits for the store
- faster support team workflows,
- less manual searching for information,
- simpler access to product and module knowledge,
- an additional support channel for customers,
- more consistent answers compared to generic AI chat,
- can be deployed in both Admin and storefront,
- a foundation for expanding additional AI features on the same knowledge base.
How using the module looks
In Admin Panel
A staff member opens the AI tab, enters a question, and gets an answer related to the store offer. They can ask, for example, about:
- a product,
- a module,
- a feature,
- documentation,
- use cases,
- basic differences between solutions.
On the storefront
A customer can use the chat as a simple informational assistant. Instead of clicking through many pages, they receive an answer faster and can navigate to the right product.
Who this module is most useful for
- store owners who want structured access to knowledge,
- support and sales teams,
- stores with a technical offer,
- companies with ready product content, manuals, and FAQ,
- stores that want to roll out AI support step by step.
Requirements
To run the module you need:
- Magento 2,
- an active OpenAI account,
- a prepared AI knowledge base for the store,
- server internet access,
- correct configuration in Magento Admin.
Installation and configuration
This document explains how to launch the AI Product Support module in a Magento 2 store and how to prepare it for use.
It is written for a store that wants to deploy a ready module and start using AI chat without diving into code level details.
It is best to treat this module as a user facing layer. First the store prepares the knowledge, and only then exposes it to staff or customers in the form of chat.
What to prepare before installation
Before deployment you need:
- a working Magento 2 store,
- server access,
- Composer,
- an OpenAI account,
- a prepared AI knowledge base for the store,
- server internet connectivity.
The most important practical requirement is simple: the module will answer well only when the store has prepared content that the AI can use.
If the store already uses Kowal AI Product Feed, it can serve as the content preparation and organization layer for AI. On that module page, this is described as building a structured and up to date knowledge base for AI systems. AI Product Support is the natural next step, using that knowledge in a conversation with the user. Source: Kowal AI Product Feed for OpenAI Vector Store
Module installation
The module is installed via composer.
Example installation process:
composer config repositories.ai.product.support vcs https://github.com/kowalco/ai-product-supportcomposer config --global --auth github-oauth.github.com composer require kowal/module-ai-product-supportbin/magento module:enable Kowal_AiProductSupportbin/magento setup:upgradebin/magento cache:clean If the store runs in production mode, after installation also run:
bin/magento setup:di:compilebin/magento setup:static-content:deploy -fbin/magento cache:cleanWhere to find configuration
After installation, you can find module settings here:
Stores > Configuration > Kowal AI > AI Product Support
How to configure the module
1. Enable chat in Admin Panel
In the General section enable:
Enable Chat
After saving, an AI tab will appear in the Admin Panel.
2. Decide whether to enable storefront chat
If you want customers to use the chat, enable:
Enable Frontend Chat
If the module is only for your internal team, keep storefront disabled.
3. Set question length
Field:
Maximum Question Length
defines the maximum message length a user can send. This helps keep things tidy and limits overly long, unclear requests.
4. Set the default store view
If your store runs multiple languages or has multiple store views, you can select:
Default Store View
This improves admin usability and helps start with the correct context.
5. Enable or disable manual store switching
Field:
Allow Store Switcher
controls whether an admin user can switch the store view directly in the chat popup.
This is useful when one team handles multiple store views.
6. Enable technical logging during rollout
Field:
Log Chat Requests
is worth enabling during configuration and testing. It makes it easier to verify that the module works correctly. After production rollout, you can decide whether to keep logging enabled.
AI connection settings
In the OpenAI section, configure the core elements required for the module to work.
OpenAI API Key
This is the access key for the AI service. Without it, the module cannot load the models list, read the knowledge source, or send questions.
Response Model
This is the model responsible for generating responses.
The models list loads from the API. If options do not appear after saving the key, use the refresh button for the models list.
Vector Store
This is the selected knowledge source for the module.
In simple terms:
- it is where the content used by the chat is stored,
- the module searches for answers there,
- if you select the wrong source, answers will be weak or incomplete.
If the store already has a structured AI knowledge base, you select that source here.
The knowledge sources list can also be refreshed in the configuration.
Maximum File Search Results
This setting defines how many supporting materials the module considers when preparing a single answer.
In practice it affects:
- answer quality,
- performance,
- AI usage cost.
A good starting value is a mid range setting, for example 6.
Storefront security
If the chat is enabled on the storefront, set security limits right away.
In the Frontend Security section you will find:
Requests Per MinuteRequests Per HourMinimum Submit Delay
These settings help limit:
- excessive message frequency,
- bot abuse,
- unnecessary resource usage and costs.
Additional system prompt
In the Prompting section you can fill in:
System Prompt Template
This field is optional. In most deployments you can leave it empty and use the module default.
Overriding makes sense only if the store wants a custom response style or additional communication rules.
How the module works from the user perspective
In the Admin Panel
After enabling the module, the user sees an AI tab on the right edge of the screen. Clicking it opens the chat panel.
In the panel you can:
- enter a question,
- receive an answer,
- see product cards if the answer relates to a specific product,
- work without page reloads.
On the storefront
If chat is enabled, a similar tab appears on the storefront. The customer can ask a question without leaving the product page or listing.
How to think about this kind of rollout
The most practical model looks like this:
- the store organizes content and prepares the knowledge base,
- the administrator selects that knowledge base in the module configuration,
- the user uses the chat,
- answers are generated based on content prepared by the store.
This approach produces better results than enabling chat alone without prepared data.
Recommended starting configuration
Admin Panel
Enable Chat=YesEnable Frontend Chat= depends on the projectMaximum Question Length=1000Allow Store Switcher=Yesfor multiple store viewsLog Chat Requests=Yesduring testingMaximum File Search Results=6
Storefront
Requests Per Minute= a conservative starting valueRequests Per Hour= a value matched to store trafficMinimum Submit Delay= at least1
What to verify after rollout
After saving configuration, run a simple test:
- verify the AI tab appears in Admin Panel,
- open the popup and send a question,
- confirm the answer is displayed correctly,
- check whether the products section appears for product specific questions,
- if storefront is enabled, verify the tab works on the storefront as well.
Most common issues
Models or knowledge sources are not visible
Most often this means:
- missing or invalid OpenAI key,
- no server internet connectivity,
- cache not cleared after changes.
Chat answers are weak or off topic
The most common cause is not the module itself but the quality of the prepared knowledge base. If content is incomplete, outdated, or too thin, answers will also be weaker.
Before evaluating the module, verify:
- whether product descriptions are meaningful and complete,
- whether FAQ and documentation are up to date,
- whether the knowledge base contains content users actually need,
- whether the selected knowledge source is correct for the store.
Storefront blocks the user too quickly
Then review the rate limits and submit delay settings in the Frontend Security section.
Short deployment checklist
- Install the module via
composer. - Enable the module and run
setup:upgrade. - Configure OpenAI connectivity.
- Select the correct knowledge source.
- Enable chat in Admin.
- Optionally enable chat on the storefront.
- Set security limits.
- Test with questions about real products and store content.
Short copy for the Installation and configuration section
The module is installed via composer and configured under Stores > Configuration > Kowal AI > AI Product Support. The administrator enters OpenAI access data, selects a model and a store knowledge source, then enables chat in the Admin Panel and optionally on the storefront. You can also set security limits and adjust core module parameters.








