Skill

Analyze AI/Tech News for Briefings

Free lite skill that collects AI/tech news from RSS feeds and ranks it with Claude-powered sentiment and impact scoring.

Works with github

81
Spark score
out of 100
Updated 20 days ago
Version 14.1.0
Models
claude

Add to Favorites

Why it matters

Automate the collection, deduplication, and analysis of AI and technology news from multiple RSS feeds to generate concise, structured briefing cards.

Outcomes

What it gets done

01

Collect news from 4+ RSS feeds (TechCrunch, The Verge, Ars Technica, Hacker News).

02

Deduplicate articles across sources to avoid redundant coverage.

03

Rank articles by importance and generate summaries with sentiment, impact scores, and industry tags.

04

Output formatted briefing cards for quick review and distribution.

Install

Add it to your toolbox

Run in your project directory:

curl -fsSL https://spark.entire.vc/get/ag-news-sentiment-engine | bash

Overview

News Sentiment Engine (Free)

A free lite news-briefing skill that pulls AI/tech articles from RSS feeds and outputs ranked, deduplicated summaries with Claude-powered sentiment and impact scores. Use for a quick multi-source AI/tech news briefing; not for investment-grade analysis, which the paid full version covers separately.

What it does

Collects and analyzes AI/tech news from four or more RSS feeds - TechCrunch, The Verge, Ars Technica, and Hacker News - described as the open-source "lite" version of a production news-analysis engine that has processed 50+ articles daily since 2026. It deduplicates overlapping coverage across sources, ranks articles by industry impact, and outputs a structured briefing card. For each article the format is fixed: a title, source, and publish date; a 2-3 sentence summary; industry tags such as AI, Semiconductor, or Cloud; a positive/negative/neutral sentiment label; a 1-5 impact score; and a one-sentence commentary giving an industry perspective. The documented usage pattern asks it to collect the latest AI/tech news, rank the top five by importance to the industry, and output the structured briefing card with all of those fields per article.

When to use - and when NOT to

Use it when preparing a concise AI or technology news briefing from multiple RSS sources, needing ranked article summaries with sentiment, tags, and impact scoring, monitoring industry changes across product launches, policy moves, and infrastructure shifts, or deduplicating overlapping coverage before writing a daily or weekly briefing. Do not treat its sentiment and impact scores as authoritative market or policy analysis - the source frames them explicitly as briefing aids, and recommends cross-checking outputs against original article sources before publication or investment use. A separate paid "Full bundle" ($29 one-time, sold externally) adds investment-grade analysis - portfolio impact scoring, options flow correlation, earnings catalyst detection - and Telegram auto-delivery, none of which this free lite skill includes.

Inputs and outputs

Input is RSS feed content from the four named sources or equivalent feeds. Output is a ranked, deduplicated briefing card in the fixed structured format above. An optional local setup clones a third-party Node project:

git clone https://github.com/tellmefrankie/news-engine
cd news-engine
pnpm install
cp .env.example .env
### Requires: ANTHROPIC_API_KEY
pnpm dev -- --collect-only

The source explicitly warns to review and pin that third-party repository yourself, and not expose API keys to an unreviewed checkout.

Integrations

Pulls from TechCrunch, The Verge, Ars Technica, and Hacker News RSS feeds, and uses Claude (via an ANTHROPIC_API_KEY) for sentiment analysis; the free tier requires no paid APIs beyond that key. The optional local setup runs on pnpm.

Who it's for

Analysts, editors, and researchers who need a fast, cross-source AI/tech news briefing with sentiment and impact scoring, and who understand the free tier is a lighter, non-investment-grade version of a paid production engine.

FAQ

Common questions

Discussion

Questions & comments · 0

Sign In Sign in to leave a comment.