Read RSS feeds

​

Purpose

• Monitor news sites or blogs for new articles on specific topics.
• Aggregate content from multiple sources into a single data stream.
• Trigger workflows based on new content publications.
• Feed article links and summaries into other blocks for further analysis or processing.

​

Configuration

1. Add a URL:
   • This is the input field where you add the URL of the RSS feed (e.g., https://example.com/feed.xml).
   • You can add multiple RSS feed URLs to the list. The block will process each one.
   • Use the + Add button for single entries. To add multiple URLs at once, use the “Add a list of pages” option from the dropdown menu next to the + Add button.
   • You can also use variables from a preceding Open datasets block to dynamically provide RSS feed URLs.
2. Managing URLs:
   • Toggle ON/OFF: Each RSS feed URL has a toggle switch to enable or disable it for a given run without deleting it.
   • Edit/Delete: You can edit or delete individual URLs from the list.
3. Advanced Options:
   • Click the Advanced link (typically in the bottom right) to access more settings.
   • Maximum number of posts to fetch from each RSS feed: This crucial parameter lets you control how many of the latest items are retrieved from each feed. If you set it to 10, the block will only fetch the 10 most recent posts, even if the feed contains more. This is useful for managing the amount of data processed in each workflow run.

Screenshot: Read RSS feeds block configuration panel

​

Output Data Structure

• title: The title of the article or post.
• link: The direct URL to the full article.
• description or summary: A short summary or the full content of the post.
• pubDate: The publication date of the post.
• author: The author of the post (if available).
• category: The category or tags associated with the post (if available).

​

Example: Summarize New Blog Posts

1. Read RSS feeds Block:
   • URL: Add the URL of a blog’s RSS feed, e.g., https://my-favorite-blog.com/rss.
   • Advanced ➙ Maximum number of posts…: Set to 5 to only get the latest 5 posts.
2. LLM Completion Block (added next):
   • Enter your prompt:
     Copy

     Summarize the following blog post in one tweet (under 280 characters).
     Title: {{title}}
     Description: {{description}}
     

   • Output format as: Variable
   • Variable name: tweet_summary

• Result: The workflow fetches the 5 latest posts, and for each one, it generates a tweet-sized summary. The tweet_summary variable (along with title, link, etc.) is now available for each post and can be used in subsequent steps, such as passing it to an Extract Data block to structure a final report.

​

Key Considerations

• Valid RSS URL: Ensure the URL you provide is a direct link to a valid RSS or Atom feed (usually ending in .xml or /feed). It is not a link to a regular webpage.
• Direct Parsing: While a browser is used technically, this block does not interact with visual webpage elements. It directly parses the structured data from the feed file.
• Data Source: This block acts as a data source, similar to Open datasets. The blocks that follow it will execute for each post fetched from the feed(s).