AI-Curated Daily News Digest

Turn RSS overload into a ranked, summarized daily briefing using AI deduplication

RSS AI news aggregation automation content-curation shortcuts summarization

Problem it solves

High-volume RSS feeds create information overload, making it impossible to identify what actually matters each day without spending hours reading.

Best for

Content creators, journalists, or knowledge workers who monitor multiple sources daily and need a focused, actionable briefing fast.

Not ideal for

Topics too niche for multi-source signal to be meaningful, or use cases requiring real-time monitoring rather than end-of-day digests.

Overview

Why this framework exists

Rather than reading dozens of individual RSS items as they arrive, this workflow aggregates all feeds for a given topic area, uses AI to deduplicate overlapping stories, and ranks remaining items by how many independent sources covered each one — a proxy for importance. The user then makes a fast human selection from the top-ranked stories, and AI summarizes the full text of each chosen item. The result is a structured daily briefing produced in minutes instead of hours, with importance signal baked in before the human ever reads a word.

Core principles

5 total

Coverage frequency across independent sources is a reliable proxy for story importance
Human judgment should be applied to ranked candidates, not raw firehose
AI handles deduplication and summarization; humans handle final selection and framing
A daily snapshot rhythm beats continuous monitoring for most knowledge workers
Automating the aggregation layer frees cognitive bandwidth for interpretation

Steps

7 steps

Collect and organize your RSS feed URLs by topic
Gather all RSS feeds relevant to your coverage area into a single list or folder. Group them by topic (e.g., tech, pop culture, industry) so the AI has context for categorization.
Pro tipUse a service like Feedbin to manage subscriptions centrally; it makes feed URL management and API access straightforward.
Filter articles to the current 24-hour window
Configure your shortcut or script to pull only articles whose publish date matches today. This eliminates stale items and keeps the ranked list relevant.
WarningSome RSS feeds use unreliable or missing publish dates. Build a fallback that filters by the feed's item order if dates are absent.
Pass all article titles and sources to AI for deduplication and ranking
Send the full list of today's article titles along with their source names to an AI prompt instructing it to identify unique stories, ignore duplicates, and rank stories by the number of independent sources that covered each one.
Pro tipExplicitly tell the AI: 'If a story appears in multiple sources, treat higher coverage count as higher importance.' This single instruction produces surprisingly accurate prioritization.
WarningDo not ask the AI to summarize at this stage. Keep this step to ranking only, or the output becomes unwieldy and expensive.
Review the AI-ranked story list and apply your own editorial judgment
Read through the top-ranked stories the AI surfaces — typically 8 to 12 items. Apply your own knowledge of your audience to mentally note which ranked stories are genuinely relevant versus algorithmically prominent.
WarningDo not skip the human review step. AI ranking by coverage frequency will occasionally surface a viral but irrelevant story. Your editorial judgment is the quality filter.
Select your target number of stories
Choose the specific stories you want in your briefing — for example, 5 from the top 10. This selection step is the key human contribution; make it deliberate.
Pro tipBuild a Choose From List step in Apple Shortcuts so selection is a tap-and-go interaction rather than a copy-paste workflow.
Have AI retrieve full text and generate a summary for each selected story
Pass each selected article's URL or headline back to AI with an instruction to summarize the full article in 3 to 5 sentences. This is the core value-generation step.
Pro tipPrompt the AI to include the key fact, the implication, and one contrasting perspective where available. It produces more usable summaries than a plain 'summarize this' prompt.
Compile summaries into your structured daily briefing
Assemble the AI summaries into your chosen output format — a show script, email digest, internal briefing doc, or slide deck. Label each story with its source count for transparency.

Checklist

Saved in your browser

Collect all RSS feed URLs for your topic area
Configure script or shortcut to filter articles to the last 24 hours only
Pass article titles and sources to AI with explicit deduplication and ranking instructions
Review AI-ranked list and identify the stories you want to cover
Select your target number of stories from the ranked list
Have AI retrieve full text and generate a summary for each selected story
Assemble summaries into your briefing, script, or digest format
Review for accuracy before publishing or using

Examples

2 cases

Daily Tech Podcast Segment for Members

The host runs a daily show segment for paying members covering top tech news. He built an Apple Shortcut that pulls RSS feeds from multiple tech publications, filters for today's articles, passes them to AI with instructions to deduplicate and rank by coverage frequency, then presents a Choose From List of the top ten stories. He selects five, the Shortcut pulls full article text for each, AI summarizes them, and he has a structured show script ready to read in minutes.

OutcomeA five-story daily tech briefing is produced in a fraction of the time manual curation would take, with importance signal validated by multi-source coverage data.

Mac Power Users podcast

Morning Newsletter Briefing for Knowledge Workers

A marketing analyst monitoring competitive intelligence across fifteen industry blogs and news sites uses the same pattern: all feeds flow into a daily aggregation script, AI deduplicates and ranks by source coverage, the analyst selects three to five stories of relevance to their company, AI summarizes each, and the output becomes a morning Slack post to the leadership team — produced before 9am with no manual reading required.

OutcomeLeadership receives a daily competitive briefing in Slack without requiring the analyst to spend ninety minutes reading individual articles.

Common mistakes

3 traps

Skipping the human selection step

Letting AI automatically select which stories to summarize removes the editorial judgment that makes the digest useful to your specific audience. The ranking surfaces candidates; you make the final call.

Asking AI to summarize at the ranking stage

Combining deduplication, ranking, and summarization into a single prompt produces long, expensive, and often inconsistent output. Keep the ranking call and the summarization call as separate steps.

Relying solely on coverage frequency for importance

High coverage frequency can reflect virality or controversy rather than genuine relevance to your audience. Always apply domain knowledge during the human review step before selecting stories.