A practical automation for turning RSS feeds into an AI-generated daily digest. n8n handles the schedule, the feed collection, and the message delivery, while the LLM turns each new item into a short summary that is easier to scan than a raw feed reader. It is a good fit for anyone who wants a lightweight news workflow without building a custom backend.
What it does
- Fetch new items from multiple RSS feeds on a schedule
- Summarize each item with an LLM in two or three sentences
- Aggregate the summaries into a single digest message
- Push the digest to Slack, Lark, email, or another webhook target
Why it is useful
This workflow removes the repetitive part of reading the same sources every day. Instead of opening a dozen blogs and news sites one by one, you get a compact brief in the channel you already use for work.
It is also easy to adapt. The same flow can be used for competitor monitoring, industry tracking, research collection, or creator inspiration. Once the skeleton is in place, the only thing that changes is the list of feed URLs and the destination channel.
Prerequisites
- An n8n instance, either cloud-hosted or self-hosted with Docker
- An LLM API key
- A Slack or Lark webhook, if you want to push the digest to a team channel
Install / configure
Minimal self-host example:
docker run -it --rm -p 5678:5678 -v ~/.n8n:/home/node/.n8n n8nio/n8n
Typical node chain: Schedule Trigger → RSS Read → loop over items → HTTP Request to the model → Set / merge summaries → Slack or Lark node.
A simple model prompt might look like this:
{
"model": "deepseek-chat",
"messages": [{ "role": "user", "content": "Summarize the following article in three sentences: {{ $json.content }}" }]
}
How to use
- Add the RSS feed URLs you want to monitor.
- Connect the model API key in the HTTP Request step.
- Configure the final push node with your webhook or channel credentials.
- Activate the workflow and confirm that the scheduled run produces a digest.
Use cases
- Daily industry and competitor intelligence
- Internal team news briefings
- Content research collection for creators and editors
- Lightweight monitoring for niche topics
Notes
- If you use an overseas model provider, the host running n8n must be able to reach that API.
- With many feeds, watch call frequency and cost, especially if the digest runs several times per day.
- It is worth adding deduplication so repeated stories do not show up every time the workflow runs.