> ## Documentation Index
> Fetch the complete documentation index at: https://docs.a2v2.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Add Links (Websites)

> Train your agent on the content of a web page by adding its link.

Add a **Links** source to train your agent on content that already lives on the
web — your marketing pages, help center, documentation, or any public website URL.
The **Links** tab is titled "Add website links to train your AI Agent."

## Prerequisites

* An agent (see [Create an agent](/building/create-agent)).
* A publicly reachable URL (pages behind a login can't be crawled).
* Enough credits to process the page (see [Credits](/concepts#credits)).

## Add a link

<Steps>
  <Step title="Open the Knowledge Base">
    In the agent sidebar, go to **Expertise → Knowledge Base**.
  </Step>

  <Step title="Open the Links tab">
    Select **New Knowledge Item**, then open the **Links** tab.
  </Step>

  <Step title="Enter the URL">
    Type the full address of the website you want to index (for example,
    `www.example.com`).
  </Step>

  <Step title="Add the link">
    Select **Add link**. A2V2.ai fetches the page, extracts its text, and indexes
    it. The source shows **Processing** until it's done.
  </Step>

  <Step title="Wait for Completed">
    When the status turns **Completed**, the page content is live in your agent.
  </Step>
</Steps>

<Frame>
  <img src="https://mintcdn.com/a2v2ai/ItYQQYTHsiiRmr-t/images/knowledge-base/links.png?fit=max&auto=format&n=ItYQQYTHsiiRmr-t&q=85&s=4b1e79d24fa224b3d547bbf63017a012" alt="The Links tab of New Knowledge Item with a website URL entered." width="1440" height="900" data-path="images/knowledge-base/links.png" />
</Frame>

## Options & limits

| Item        | Detail                                                                                              |
| ----------- | --------------------------------------------------------------------------------------------------- |
| **Input**   | A single public URL per source                                                                      |
| **Access**  | The page must be publicly reachable (no login walls)                                                |
| **Content** | Visible text is indexed; images and video are not                                                   |
| **Updates** | Re-crawl by [reprocessing](/knowledge-base/manage-sources#reprocess-a-source) when the page changes |

<Note>
  A website source captures the page as it is *at processing time*. If you update
  the page later, reprocess the source so the agent learns the new content.
</Note>

## Tips

<Tip>
  * Point at content-rich pages (docs, FAQs, articles) rather than thin landing
    pages.
  * Add the most important pages individually so each is indexed cleanly.
  * For content you control, an [uploaded file](/knowledge-base/upload-files) or
    [Q\&A pair](/knowledge-base/qa) often gives more predictable results than a
    crawl.
</Tip>

## Troubleshooting

<AccordionGroup>
  <Accordion title="The crawl returned little or no content">
    The page may render its content with JavaScript, or hide it behind a login or
    cookie wall. Try a more static URL, or paste the key content as an
    [uploaded file](/knowledge-base/upload-files) or [Q\&A](/knowledge-base/qa).
  </Accordion>

  <Accordion title="The source failed">
    Confirm the URL is correct, public, and reachable in a normal browser, then
    [reprocess](/knowledge-base/manage-sources#reprocess-a-source).
  </Accordion>
</AccordionGroup>

## Related

<CardGroup cols={2}>
  <Card title="Upload a file" icon="file-arrow-up" href="/knowledge-base/upload-files" />

  <Card title="Add YouTube" icon="youtube" href="/knowledge-base/youtube" />

  <Card title="Manage & reprocess sources" icon="arrows-rotate" href="/knowledge-base/manage-sources" />
</CardGroup>