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

# Connect an RSS or Atom feed

> Keep the articles from a public RSS or Atom feed searchable by your AI agents

Connect a public RSS or Atom feed and Raily checks it for new entries every hour. Each entry becomes a searchable article in one **feed**, which you can add to any endpoint.

## Before you start

* You need the public URL of the feed, for example `https://example.com/feed.xml`. Raily supports RSS 2.0 and Atom. RSS 1.0 (RDF) feeds are not supported.
* The feed has to be reachable from the internet without a password. URLs with a username or password in them are rejected.
* The feed needs at least one entry. Raily refuses an empty feed at the connection test.

## Connect the feed

**1. Open the source catalog**

In **Search Rail**, click **Connect source**.

**2. Choose RSS / Atom**

On **Connect a data source**, click the **RSS / Atom** card.

**3. Name the source**

In **Name**, type something you'll recognize later, such as `Daily news feed`.

**4. Enter the feed URL**

Paste the address into **Feed URL**. If you leave out `https://`, Raily adds it.

**5. Test the connection**

Click **Test connection**. Raily fetches the feed and shows its title, whether it's RSS or Atom, and how many current entries it found.

**6. Connect**

Click **Connect RSS / Atom**. The button stays off until the test passes. If you change the URL after testing, run the test again.

## What happens next

Raily opens the new source and starts the first sync right away. You don't need to do anything while it runs.

1. Raily reads the entries currently in the feed (up to 500) and indexes them.
2. When the first sync finishes, Raily profiles the content and generates the search config for the feed. This usually takes under a minute.
3. The feed card shows **Last sync** with the time and counts of added, updated and removed entries.

To make the feed searchable, add it to an endpoint from **Searched by** on the feed card. See [AI endpoints](https://docs.raily.ai/concepts/ai-endpoints.md).

After that, Raily checks the feed every hour. To check right now, click **Resync** on the feed card.

### Older entries stay searchable

Most feeds only list their latest entries. When an entry drops out of the feed, Raily keeps it. Your archive grows with every sync instead of shrinking to the feed's current window. If an entry changes in the feed, Raily updates the stored copy rather than adding a duplicate.

Have older articles that were never in the feed? The source page says **Have an article archive? Contact us**. Send us an export and we'll import your history into the same source.

## What gets indexed

Each feed entry becomes one article. Raily keeps the fields the feed provides:

| Field | What it holds |
| - | - |
| Title | The entry's headline |
| Text | The entry's summary or full content, with HTML cleaned up |
| Link | The original article, used as the link on result cards |
| Published date | When the entry was published. Readers can ask for "this week" or "in March" |
| Authors | The entry's authors, when the feed lists them |
| Categories | The feed's own categories or tags, when present. Readers can filter on them |
| Language | The language of the entry |
| Image | The entry's own image. If it has none, Raily uses the article page's preview image, then the feed's logo |

Raily also reads each article and adds the people, organizations and topics it mentions, so a reader can ask for "articles about the central bank" and get matching entries even when the feed has no categories.

To review or change how each field is used in search, open the feed in the [Data Catalog](https://docs.raily.ai/integrations/data-catalog.md).

## Edit or remove the source

Open the source and click the pencil (**Edit connection**). You can rename the source or turn it on and off. The **Feed URL** can't be changed, because it identifies the source. To follow a different feed, connect a new source.

To remove it, click **Delete source** on the same page and confirm.

## Next steps

> [card] [Create an AI endpoint](https://docs.raily.ai/concepts/ai-endpoints.md)
>
> Let agents search this source

> [card] [Data Catalog](https://docs.raily.ai/integrations/data-catalog.md)
>
> Review the fields agents filter on


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.