zenitlab

zenitlab / ai-hot-radar

Public

React19+NestJS 实现的ai情报站，集成20+RSS信源，包含twitter、hacker news等以及7个搜索信源，数据清洗合并，AI评分展示后展示AI圈最有价值的消息

18
0
85% credibility
Found May 24, 2026 at 20 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
TypeScript
AI Summary

AI Hot Radar is a real-time news aggregator designed specifically for AI practitioners. The system continuously monitors over 20 sources—including Twitter, major AI blogs, academic papers, Chinese tech news, and video platforms—then uses AI to read, score, and filter every article. It surfaces only what's genuinely important, generates a daily morning summary organized into sections like 'Model News' and 'Industry Trends', and lets you subscribe to specific topics you care about. The tool also offers ways to connect it to AI assistants and RSS readers so the news fits naturally into your existing workflow.

How It Works

1
🔍 You discover AI Hot Radar

You hear about a tool that watches AI news for you and only shows what's worth reading.

2
⚙️ You set up the app

You install the app on your computer and connect it to an AI service so it can read and judge the news for you.

3
🌐 The app starts watching the internet

Behind the scenes, the app continuously checks Twitter, news sites, blogs, and video platforms for AI news.

4
✨ AI reads and scores every article

For each piece of news, your AI helper decides if it's real, how important it is, and whether AI people should care about it.

5
You explore in different ways
⭐
Browse curated news

See only the highest-quality AI articles, sorted by importance

📰
Read your daily digest

Every morning you get a summary of the most important AI news

🔔
Track your keywords

Add topics like 'Claude' or 'Cursor' and get notified when they appear

6
🔗 You connect it to your AI assistant

You install a special skill so your AI assistant can look up AI news for you when you ask.

🎉 You stay informed without the noise

Instead of checking dozens of websites, you have one place that finds, filters, and summarizes what's actually important in the AI world.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 20 to 18 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is ai-hot-radar?

AI Hot Radar is a real-time news aggregator built for AI practitioners. It continuously pulls from 20+ sources — Twitter, Hacker News, RSS feeds from AI labs, Chinese tech media, and search engines — then runs content through a two-stage AI scoring pipeline to surface only what matters. The frontend is React 19 with a clean dashboard; the backend is NestJS with Prisma and SQLite. Results land in three views: a curated feed, a daily AI-generated digest, and a keyword monitoring system with live push notifications. The system also exposes itself as a Skill for AI coding assistants, an RSS feed for news readers, and a REST API for bots and automation pipelines.

Why is it gaining traction?

The hook is the two-stage scoring: a cheap AI model pre-filters noise, then a full scorer grades survivors. This keeps costs down while still catching quality content across sources that vary wildly in signal-to-noise ratio. The daily digest at 08:00 Beijing time is genuinely useful — six sections covering models, products, industry, community, and papers, all auto-generated. The keyword expansion feature is clever: it automatically finds synonyms and judges authenticity, so you get relevant hits without babysitting filters. The Agent integrations are the real differentiator — you can pipe curated content into Claude Code, Feedly, or a Feishu bot with zero custom code.

Who should use this?

AI researchers and developers who want a personalized news radar without manually checking 15 different sources. Startup founders tracking competitive AI landscape. Technical content creators who need a daily briefing. Teams that want to pipe AI news into internal Slack or WeChat bots. It's less suited for non-technical users — the setup requires an OpenAI-compatible API key and running two terminal processes.

Verdict

At 18 stars with minimal test coverage and a self-hosted model, this is a solid personal tool or internal project — not production-ready SaaS. The credibility score of 0.85% reflects that maturity gap. If you're comfortable self-hosting and want a customizable AI news pipeline, clone it and tune the scoring thresholds. If you want plug-and-play, wait for a hosted version.

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.