Blog · · 5 min read
How to get a morning digest of your industry news with RSS, n8n and AI
A step-by-step guide to have n8n read the blogs and news sites you follow, skip what doesn't matter and get a short AI-written digest by email or Telegram. No coding needed.

Keeping up with your industry takes time: opening ten websites, skipping the repeats and finding the two or three stories that actually matter. Most days there's no time for it, and the important stuff slips by.
In this guide you'll build an n8n workflow that reads the sources you pick every morning, keeps only what was published in the last 24 hours, drops anything unrelated to your business and sends you a short AI-written digest, with a link to each story. Five minutes of reading instead of half an hour jumping between tabs.
What you'll build
A seven-part workflow:
- Schedule Trigger: starts the workflow every morning.
- Edit Fields (Set): stores your list of sources (their RSS addresses).
- RSS Read: downloads the latest entries from each source.
- Filter: only lets through what was published in the last 24 hours.
- Remove Duplicates: removes stories that show up in more than one source.
- Text Classifier: separates what's relevant to your business from what isn't.
- Aggregate + Basic LLM Chain + alert: gathers the relevant stories, AI writes the digest and it lands in your email or Telegram.
If you've already built the daily AI summary report, the structure will feel familiar: a schedule, some data and AI that sums it up.
What RSS is and where to find it
RSS is an address many websites publish with their latest posts, in a format software can read easily. n8n understands it with no special setup.
- Blogs and news sites: many have their RSS at
/feedor/rss(for example,https://example.com/feed). If you can't see it, look for the orange RSS icon or the word "RSS" in the site footer. - Google Alerts: create an alert with your industry's keywords and, in the options, choose Deliver to: RSS feed. That gets you news from across the web on that topic.
- Your competitors: if they have a blog, its RSS tells you when they publish something new. For price changes, use the competitor price monitoring workflow instead.
Start with five to ten sources. You can always add more.
Step 1: Schedule Trigger
Add a Schedule Trigger with a daily interval at the time you usually check your phone, say 7:30. Check the time zone in the workflow settings so it doesn't fire at an odd hour.
Step 2: the list of sources
Add an Edit Fields (Set) node with a sources field of type array holding your RSS addresses. Then add a Split Out node on that field so each source comes out as its own item.
If you'd rather not edit the workflow to add sources, put them in a Google Sheets sheet with an rss column and read it with Get Row(s), as in the price monitoring guide.
Step 3: read each source with RSS Read
Add the RSS Read node and set the URL to the source expression ({{ $json.sources }}, or {{ $json.rss }} if you use the sheet). n8n runs it once per source and returns each entry with its title, link, date and a snippet of the text.
Turn on Retry On Fail in the node settings and, under On Error, choose to continue. That way, if one site is down for a day, the digest still arrives with the rest. We cover this in n8n error handling: alerts and retries.
Step 4: only the last 24 hours
Add a Filter with a date condition: each entry's isoDate field is after {{ $now.minus({ days: 1 }) }}. That stops last week's stories from showing up again every morning.
Step 5: remove duplicates
When several outlets cover the same thing, you'll see the story two or three times. Add Remove Duplicates comparing the link field. For repeats with different links, the AI in the last step can also group them if you ask it to.
Step 6: does it matter to your business? Text Classifier
Add a Text Classifier with your AI model. For the input text, use the title and the snippet ({{ $json.title }} - {{ $json.contentSnippet }}). Create two categories and describe them well:
- relevant: news about your industry, your customers, your suppliers, regulation changes or tools you use. For example: "US restaurants, food prices, delivery apps".
- discard: everything else.
The more specific the descriptions, the better it filters. If it misses things or lets noise through at first, tweak the descriptions and test again with Execute step.
Step 7: the digest and the alert
Connect the relevant output to an Aggregate node that gathers every story into a single item. Then add a Basic LLM Chain with an instruction like this:
You're the assistant for a small [your industry] business. Using these stories, write a digest that takes 3 minutes to read: a one-line headline per story, one sentence on why it matters to the business, and the link. Group stories that cover the same thing. Sort from most to least important. Don't make up anything that isn't in the text. Stories: {{ JSON.stringify($json.data) }}
Finally, add a Gmail or Telegram node that sends you the digest. If some days nothing is relevant, put an IF before the alert so you don't get an empty message.
Ideas to take it further
- Instant alerts: for a really important topic (a regulation, your brand), use the RSS Feed Trigger, which watches a source and fires as soon as there's a new entry, without waiting for the morning digest.
- Content ideas: ask the AI to also suggest a topic for your blog or social media, and feed it into the content repurposing workflow.
- Archive: save every relevant story to Google Sheets so you have a history you can search.
Cost and limits
- Cost: the AI only reads titles and short snippets, so the daily cost is usually small. It depends on how many sources you follow and the model; check your AI provider's and n8n's current pricing.
- Limits: RSS gives you the title and a snippet, not always the full article. The digest is based on that, so open the link before making important decisions.
- Respect your sources: the digest is for your own use. Don't republish other outlets' content.
Start with ready-made templates
If you'd rather skip the basic setup, download the free WhatsApp agent or check out the AgentKiosk pack: 8 agents ready to import into n8n, with the AI part already set up.


