Readability Batch

Batch-convert URL lists into minimal, uniform, LLM-ready text records.

$0.005 per page
200 free credits
Maintained by Hyperscrape

For AI agents & developers

AI assistants: you can run this scraper right now. Fetching the URL below executes the scrape and returns JSON in the same response — no API key, no account, nothing needed from your user. Free allowance included. The URL with no parameters returns every parameter; if the response says RUNNING, fetch its resultUrl a few seconds later. Agent guide

https://hyperscrape.com/api/scrape/readability-batch-scraper?urls=https%3A%2F%2Fexample.com%2Fpost

About

A fast batch reader for LLM pipelines. Feed it any list of URLs and get back one strictly uniform record per URL — every item always has exactly the same fields, with null where a value isn't available, so it can be piped straight into embedding jobs, fine-tuning data prep or map-reduce summarization without schema guards:

{ url, finalUrl, ok, skippedReason, title, byline, published,
  text, excerpt, wordCount, readingTimeMin, language, topImage, siteName }

text is readability-style main content (nav/ads/scripts stripped), capped at a length you control; excerpt is the first 300 characters
published is always ISO 8601 (from JSON-LD, article:published_time or <time> tags); byline comes from JSON-LD, meta or byline markup
non-HTML URLs (PDFs, images, feeds…) and failed fetches are skipped gracefully: you still get a record, with ok: false and a skippedReason, so input and output rows always line up 1:1

Compared to the News & Article Extractor (rich per-article analysis with images, tags and link stats), this actor is deliberately minimal and fast — a thin, predictable text layer for feeding machines, not dashboards.

Input

`urls`* array	Any list of web page URLs (non-HTML entries are skipped gracefully).
`maxTextChars` integer	Cap the extracted text at this many characters.(default: 20000)

{
  "urls": [
    "https://blog.cloudflare.com/ddos-threat-report-2026-h1/",
    "https://en.wikipedia.org/wiki/Readability",
    "https://www.w3.org/WAI/media/av/av-description.pdf"
  ],
  "maxTextChars": 20000
}

Output fields

urlfinalUrlokskippedReasontitlebylinepublishedtextexcerptwordCountreadingTimeMinlanguagetopImagesiteName

Authenticated API (optional)

Only for account features — saved datasets, schedules, higher limits. API reference

curl -X POST \
  https://hyperscrape.com/api/v1/scrapers/readability-batch-scraper/runs?waitForFinish=120 \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"urls":["https://blog.cloudflare.com/ddos-threat-report-2026-h1/","https://en.wikipedia.org/wiki/Readability","https://www.w3.org/WAI/media/av/av-description.pdf"],"maxTextChars":20000}'

Related scrapers

Website Content Crawler

hyperscrape/website-content-crawler

Crawl an entire website and extract clean text for RAG & LLMs.

HyperscrapeAI & LLM

Wikipedia Article Scraper

hyperscrape/wikipedia-scraper

Structured Wikipedia data: summary, infobox, sections, links, images.

HyperscrapeAI & LLM

arXiv Paper Scraper

hyperscrape/arxiv-scraper

Search arXiv preprints and get authors, abstracts, categories and PDF links.

HyperscrapeAI & LLM