I got tired of doomscrolling 7 different AI news sources every morning, so I built an agent that does it for me for exactly $0/month.
The pipeline:
- 7 RSS feeds (Hacker News, Google AI blog, Hugging Face, Lobsters, The Hacker News, Open Source blog) → a Python script on a free-tier server
- Dedup the same story hits 5 feeds; it picks the best source and drops the rest
- An LLM writes 2-3 sentence summaries of the stories that actually matter
- A cron job publishes a clean daily digest to Discord, and archives every issue to a free static site: https://apexnexus.site
What surprised me:
1. The dedup step matters more than the AI part. 60% of my "news" was the same 3 stories reblogged.
2. Self-healing is the real feature. When a webhook died, the bot just... rebuilt it. I found out days later. That changed how I think about agent reliability.
3. The whole thing runs unattended. I haven't manually hunted for AI news in weeks, and I don't miss it.
I wrote up the blueprints for each piece on the site (the self-healing webhook writeup got the most attention).
What's the most surprisingly useful automation you've built with AI? I'm looking for ideas for the next one.
[link] [comments]