> For the complete documentation index, see [llms.txt](https://docs.solarvis.co/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.solarvis.co/solarvis-modules/ai-sales-assistant/training.md).

# Training

## Purpose of This Page <a href="#purpose-of-this-page" id="purpose-of-this-page"></a>

The AI Sales Assistant answers customer questions using the knowledge configured on this page. By connecting website content and adding custom information, you define the information the assistant can use during conversations.

Website content can be refreshed automatically on a schedule or manually whenever new information becomes available, helping keep the knowledge base aligned with your latest published content.

## What You Can Do Here <a href="#what-you-can-do-here" id="what-you-can-do-here"></a>

Using the Training page, you can achieve the points:

* Build the AI Sales Assistant knowledge base from website content and custom text.
* Configure automatic website crawling and choose how frequently it runs.
* Add website domains and individual pages as training sources
* Monitor crawl progress and page processing status.
* Create and manage custom text blocks that provide additional knowledge.

## How It Works <a href="#how-it-works" id="how-it-works"></a>

Training content is organized into two source types:

* **Web Pages** - Content collected automatically from your website.
* **Text Blocks** - Knowledge entered manually.

Once a source has been successfully processed, it becomes part of the AI Sales Assistant knowledge base and can be used to answer visitor questions.

{% hint style="info" %}
**Important:** Only sources with the **Ready** status are included in the knowledge base.
{% endhint %}

### Crawl Schedule <a href="#crawl-schedule" id="crawl-schedule"></a>

The Crawl Schedule determines how often the AI Sales Assistant refreshes information collected from connected websites.

Enable automatic crawling to keep website-based training content synchronized with your published website.

Available frequencies include:

* Monthly
* Quarterly
* Semi-annually
* Yearly

If automatic crawling is disabled, website content remains unchanged until a manual crawl is started.

{% hint style="info" %}
**Note:** Automatic crawling only updates website sources. Text Blocks are updated only when they are edited manually
{% endhint %}

### Web Pages <a href="#web-pages" id="web-pages"></a>

The **Web Pages** section manages all website-based training sources.

Add one or more company websites using the **Domain** field. During crawling, the system automatically discovers pages belonging to the configured domains and processes them for use by the AI Sales Assistant.

A maximum of **200 pages** can be crawled across all configured domains.

Use the **Extra URL** field to include a specific page directly. This is useful for pages that are not easily discovered during crawling or should always be included in the knowledge base.

If website content changes before the next scheduled crawl, select **Restart Crawl** to immediately refresh the connected sources.

While crawling is in progress, the page displays:

* Total pages discovered
* Successfully processed pages
* Processing status for each page
* Source URL for every page (recrawl or delete pages one by one)

{% hint style="info" %}
Use the available language filter and update-date sorting options to quickly locate specific pages when managing large knowledge bases.
{% endhint %}

### Text Blocks <a href="#text-blocks" id="text-blocks"></a>

The **Text Blocks** section allows you to add company knowledge that is not available on your website.

This is useful for internal procedures, product information, sales guidance, FAQs, policies, or any other information that should be available to the AI Sales Assistant.

Select **Add** to create a new text block and enter the information as a separate knowledge entry. Keeping different topics in individual text blocks makes future maintenance easier.

From the Text Blocks list, you can review existing entries and organize them using the available language filter and update-date sorting options.

{% hint style="info" %}
**Recommendation:** Create smaller, topic-focused text blocks instead of large documents. They are easier to maintain and produce more targeted responses from the AI Sales Assistant.
{% endhint %}

***

## Related Pages <a href="#related-pages" id="related-pages"></a>

* [AI Sales Assistant](/solarvis-modules/ai-sales-assistant.md)
* [General Settings](/solarvis-modules/ai-sales-assistant/general-settings.md)
* [Cost Configuration](/solarvis-modules/ai-sales-assistant/cost-configuration.md)
* [Regulations](/solarvis-modules/ai-sales-assistant/regulations.md)
* [Cost Configuration Test](/solarvis-modules/ai-sales-assistant/cost-configuration-test.md)
* [Greeting](/solarvis-modules/ai-sales-assistant/greeting.md)
* [Theme](/solarvis-modules/ai-sales-assistant/theme.md)
* [AI Sales Assistant End-User Journey](/solarvis-modules/ai-sales-assistant/ai-sales-assistant-end-user-journey.md)

For more assistance, don’t hesitate to [get in touch.](https://solarvis.co/en/company/contact)


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://docs.solarvis.co/solarvis-modules/ai-sales-assistant/training.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.