RSS Feed

Point the table at an RSS or Atom feed and get a row per item — title, link, summary, publish date, author.

Most scrapers are a snapshot. A feed is a subscription: the same source, checked again, with only what is new.

What it does

• Reads a feed URL and creates a row per item
• Works with any valid RSS or Atom feed — publisher blogs, Google News queries, job boards, subreddit feeds, changelogs, podcast feeds
• Paired with a scheduled import, it becomes a standing signal source instead of a one-time pull

How to use it

1. Find the feed URL — most blogs expose one at /feed or /rss.xml
2. Open the table and click Table → Import → RSS Feed
3. Paste the URL and run it once to see the item shape
4. Map title / link / summary / date to columns and click Create Rows
5. Schedule the import so new items arrive on their own

Common use cases

• Watch a funding or news feed. New rounds become rows the day they are announced.
• Monitor a company’s blog or changelog. A shipped feature is a reason to reach out.
• Track a job board. Many boards publish a feed, which is cheaper to poll than to scrape.
• Follow a category. A Google News query has a feed URL — that URL is a category monitor.
• Feed an AI agent. Point a research agent at the summary column and have it flag only the items that matter.

Best practices

• Run once manually and confirm the item shape before scheduling it
• Dedupe on the item link — feeds re-publish edited items
• Prefer a narrow feed over a firehose; a whole-site feed buries the items you care about
• Keep the publish-date column and filter on it — an old item in a re-ordered feed looks new otherwise
• Let an AI agent triage relevance before a person reads any row
• Re-check the URL if rows stop arriving; feeds move or get retired silently

Where to next

• Manage scheduled imports — poll the feed on a schedule
• AI Agents — triage feed items automatically
• Auto Run — enrich each new row as it arrives