Turn long articles into quick notes with AI. The flow collects the latest posts from a public site, summarizes the first three, and gives you a clear list with titles, links, and short takeaways. Teams that publish newsletters or weekly digests will find it useful.

It starts on click. An HTTP Request loads the listing page, and an HTML parser grabs the article links with a CSS selector. The Split Out and Limit steps prepare up to three links. For each link, another HTTP Request downloads the page. The system pulls the page title, loads the text, splits it into smaller parts, and sends the content to an OpenAI chat model through a summarization chain. A Merge step lines up the title and summary, and a final Set node outputs clean fields.

You only need an OpenAI API key and a valid source page. Update the URL and CSS selector to match your website. Expect research time to drop when you only scan highlights instead of reading full pages. Use it for weekly roundups, competitor watch, and content planning.

• Manual run control with a click based trigger
• HTTP Request loads the article listing page
• HTML extraction collects article links using a CSS selector
• Split Out and Limit prepare a precise batch of items
• HTTP Request fetches each article page for processing
• HTML extraction pulls the page title for each article
• Recursive text splitter breaks long text into clean chunks
• AI summarization chain uses an OpenAI chat model for concise notes
• Merge aligns summaries with the correct titles and links
• Set node outputs tidy fields for title, summary, and URL

• Reduce manual reading from 90 minutes to 5 minutes for three articles
• Automate about 80% of content curation steps
• Improve title and link accuracy by pulling data directly from the page
• Scale from 3 to 50 articles by increasing the limit without extra staff
• Connect web content and OpenAI without custom code

1. Import the template into n8n: Create a new workflow in n8n > Click the three dots menu > Select 'Import from File' > Choose the downloaded JSON file.
2. You'll need accounts with OpenAI. See the Tools Required section above for links to create accounts with these services.
3. In the n8n credentials manager, open the OpenAI Chat Model node, select Credential to connect with, choose Create new credential, name it clearly, and paste your OpenAI API key from the OpenAI API page. Save the credential.
4. Open the HTTP Request node that loads the listing page and set the URL to your article index page.
5. Open the HTML extraction node for links and set the CSS selector to match your site structure. Use the node preview to confirm it returns a list of links.
6. Check the Split Out node and confirm it targets the field that holds the links. Update the field name if your extraction key is different.
7. Set the Limit node to the number of articles you want to summarize per run.
8. Open the HTTP Request node that fetches each article. If your links are relative, keep the base URL plus path. If your links are full URLs, remove the base prefix.
9. Open the title extraction node and confirm the selector returns the correct title, for example the title tag or a headline element.
10. Review the AI summarization chain and model node. Keep defaults or adjust the model and temperature as needed for your summaries.
11. Click Execute to run the flow. Inspect the Clean up node output and confirm each item includes title, summary, and URL.
12. If you get empty results, recheck the CSS selectors. If the site blocks requests, add a User Agent header in HTTP Request or slow the run with a limit. Fix any broken links before trying again.