Skip to content
Agents tracked: 442 Downloads (7d): 202M down 19.3% GitHub stars: 10.4M Releases (7d): 693 Agent status: all upSince last limit reset: Claude 28d · Codex 1d Updated Oct 21, 2026
Recipe

Build a filtered morning AI news brief with You.com, Jev, and n8n

A guest post from You.com shows how to build an n8n workflow that searches the day's AI news, filters articles with a low-cost scoring model, and posts one cited brief per topic to Slack. Topics with no material news produce no post.

The problem

Most news digests either send everything or send an LLM summary of everything, because judging relevance per article normally requires an expensive model call. The digest then arrives every morning whether or not anything happened, and readers gradually stop opening it.

Steps

  1. Install the You.com community node, get You.com and TypeSafe API keys, and connect a Slack credential and channel.
  2. Import the template and edit the Config node: topics, a description of your team, what you care about, and materialityThreshold.
  3. A Schedule Trigger runs each weekday at 8am; the You.com node searches each topic's news from the past day and skips articles seen in earlier runs.
  4. Send each article to Jev, which answers three questions in one call: real development, on topic, and how much it matters to your team.
  5. Keep articles above the threshold and group them by topic so multiple outlets covering one story become a single item.
  6. Have You.com Research write a short cited brief per topic and post it to Slack; topics with nothing material are skipped.

Results

  • As reported by the n8n blog (guest post by You.com), Jev scores each article for about $0.00004.
  • As reported by the n8n blog, a full morning run costs roughly $0.01 to $0.04, and quiet days cost less because the research step is skipped.
  • As reported by the n8n blog, sixty articles across three topics cost roughly a quarter of a cent to filter.

As reported by the source (n8n blog Verified Node Spotlight (guest post by You.com)); AgentGid did not measure these figures.

Takeaway. Run a cheap per-article filter before any expensive summarization step, and post nothing when a topic has no material news.
AgentGid's take

This suits a small team already comfortable editing n8n templates and handling API keys, since setup needs You.com, TypeSafe and Slack credentials plus a tuned materialityThreshold. The cost figures are vendor-reported (the n8n blog guest post), so check them against your own volume, and expect to adjust the threshold before the brief feels right. If you'd rather not use n8n, Flowise is Free and open source, though the template here is built for n8n.

The agent used here

Similar use cases

Guides