Filter RSS news with OpenRouter AI and write summaries to Google Sheets

## Filter and summarise RSS news with AI into Google Sheets Who's it for Anyone who has to keep up with a set of news sources - analysts, comms teams, researchers - and wants a filtered, summarised digest in a spreadsheet instead of reading every feed by hand. How it works Several RSS feeds are read in parallel and merged into one list. The first AI agent is a relevance filter: it judges each headline against your topic and returns a typed verdict with a reason, pinned to a schema so the answer is always machine-readable. Only the articles that survive go on to the second stage, which is where the cost is. That agent fetches the full article page through Browserless - an RSS item usually carries a teaser, not the piece - and writes a structured summary against its own schema. The result is shaped into rows and appended to Google Sheets. Filtering before fetching is the whole design: scraping and summarising every item in every feed would cost many times more to produce the same digest. How to set up 1. Replace the three RSS nodes with your own feed URLs - add or remove as many as you like, they all meet at the same merge. 2. Attach an OpenRouter credential to both model nodes and a Browserless credential to the fetch node. 3. Edit the prompt on Filter the news to describe the topic you actually care about. 4. Attach Google Sheets and point the last node at your sheet. Requirements An OpenRouter API key, a Browserless account for page fetching, and a Google Sheet. How to customize The relevance test is the prompt on Filter the news; the digest format is the prompt on summarise each article plus its schema. Change each prompt together with its schema, never one without the other.