Hacker News Scraper - stories, comments, jobs and full-history search

Collect Hacker News data from its two official, open APIs: the live Firebase lists (top, new, best, Ask HN, Show HN, jobs) and the Algolia search index for the whole history. One flat dataset for stories, jobs, polls and comments. US$0.30 per 1,000 items.

Modes

Mode	Use it for
lists	The live front page and other lists: top and new (500), best (200), Show HN (~150), Ask HN and jobs (~30)
search	Full-text search over every story and comment since 2006 with filters: content type, minimum points/comments, date range, sort by date or relevance. Newest-first searches can go back through any number of results
items	Specific stories or comments by ID or news.ycombinator.com link

Switch on Include comments to add the comment thread of each story (top-level comments first, up to a per-story cap) as extra items.

Input examples

Front page with comments:

{ "mode": "lists", "lists": ["top"], "maxItems": 500, "includeComments": true, "maxCommentsPerStory": 50 }

Everything mentioning a topic, last year, popular only:

{ "mode": "search", "queries": ["\"large language model\""], "tags": ["story"], "minPoints": 100, "dateFrom": "2025-10-01", "maxItems": 2000 }

Monthly "Who is hiring" style comment mining:

{ "mode": "search", "queries": ["python remote"], "tags": ["comment"], "dateFrom": "2026-10-01", "maxItems": 1000 }

Ask HN and Show HN posts:

{ "mode": "search", "tags": ["story", "ask_hn"], "sortBy": "date", "maxItems": 500 }

Output

{
  "id": 50073879,
  "type": "story",
  "title": "Gemini 4 Argon",
  "url": "https://deepmind.google/models/gemini/",
  "domain": "deepmind.google",
  "points": 294,
  "numComments": 247,
  "createdAt": "2026-10-13T16:03:16.000Z",
  "text": null,
  "parentId": null,
  "storyId": 50073879,
  "storyTitle": null,
  "author": null,
  "hnUrl": "https://news.ycombinator.com/item?id=50073879",
  "source": "list:top",
  "scrapedAt": "2026-10-13T17:14:17.109Z"
}

type is story, job, poll or comment. Ask HN, Show HN and comment bodies are in text as plain text (HTML removed, links kept). Comments have parentId and storyId.

Privacy: Hacker News usernames are not collected. The author field exists for compatibility and is always null.

Pricing

Event	Price
Item saved (story, job, poll or comment)	US$0.30 per 1,000 (US$0.0003 each)

Use Max items in total to cap the bill; comments count as items.

FAQ

Is this allowed? It uses Hacker News's official Firebase API and Algolia's public HN Search API, both intended for programmatic access.

How many results can a search return? Algolia serves 1,000 hits per query call; date-sorted searches automatically continue backwards in time, so you can pull tens of thousands of items. Relevance-sorted searches are limited to the 1,000 best hits.

How fast is it? About 60 stories per second from the lists (parallel requests) and several thousand items per second from search.

Do I need a proxy? No. It starts with direct requests and switches to a residential proxy automatically if an API rejects the IP.

Use cases: tech trend and launch monitoring, brand and competitor mentions, lead and hiring intelligence, content curation, sentiment analysis, datasets for LLM training and RAG.

En español

Scraper de Hacker News con las APIs oficiales (Firebase y Algolia): listas en vivo (top, new, best, Ask, Show, jobs), búsqueda de texto completo con filtros (puntos, comentarios, fechas) y hilos de comentarios. No se recopilan nombres de usuario. Precio: US$0.30 por cada 1.000 ítems.