---
title: "Hacker News AI Agents Digest — Weekly top discussions delivered to Telegram"
description: "Fetch this URL using the http tool:\nhttps://hn.algolia.com/api/v1/search?query=AI+agents&tags=story&hitsPerPage=10&numericFilters=points>50\n\nParse the JSON response. For each story extract:\n- Title\n- Points\n- Date\n- URL\n\nSort by points (highest first). Take top 5.\n\nThen send me a Telegram message in this format:\n\n\"🗞 Hacker News — AI Agents Digest\n\n1. [Title]\n⬆️ [Points] points · [Date]\n🔗 [URL]\n\n2. [Title]\n⬆️ [Points] points · [Date]\n🔗 [URL]\n\n(repeat for all 5 stories)\n\n💬 Most discussed: [title of #1 post]\"\n\nIf you want this automatically every week, create a routine that runs every Monday at 9:00 AM with this exact task."
canonical: "https://hub.ironclaw.com/usecases/hacker-news-ai-agents-digest-weekly-top-discussions-delivered-to-telegram"
markdown: "https://hub.ironclaw.com/usecases/hacker-news-ai-agents-digest-weekly-top-discussions-delivered-to-telegram.md"
type: "use-case"
categories: ["Personal assistant","Research","Automation"]
author: "Evgeny"
---

# Hacker News AI Agents Digest — Weekly top discussions delivered to Telegram

## Example Prompt

> Fetch this URL using the http tool:
> https://hn.algolia.com/api/v1/search?query=AI+agents&tags=story&hitsPerPage=10&numericFilters=points>50
> 
> Parse the JSON response. For each story extract:
> - Title
> - Points
> - Date
> - URL
> 
> Sort by points (highest first). Take top 5.
> 
> Then send me a Telegram message in this format:
> 
> "🗞 Hacker News — AI Agents Digest
> 
> 1. [Title]
> ⬆️ [Points] points · [Date]
> 🔗 [URL]
> 
> 2. [Title]
> ⬆️ [Points] points · [Date]
> 🔗 [URL]
> 
> (repeat for all 5 stories)
> 
> 💬 Most discussed: [title of #1 post]"
> 
> If you want this automatically every week, create a routine that runs every Monday at 9:00 AM with this exact task.

## What the Agent Does

The agent fetches the top AI agents discussions from Hacker News via the public Algolia API — no API key required. It filters only stories with more than 50 points to cut out noise, sorts by engagement, and picks the top 5. It then sends a clean Telegram digest with titles, point counts, dates, and direct links. Can be set up as a weekly routine so the digest arrives every Monday morning automatically.

![Image](https://github.com/user-attachments/assets/6567f459-9249-4aa0-abec-00bba5e3a150)

## Skills and Tools

- http
- message
- routine/cron

## Categories

- Personal assistant
- Research
- Automation

## Links

- [HTML page](https://hub.ironclaw.com/usecases/hacker-news-ai-agents-digest-weekly-top-discussions-delivered-to-telegram)