Free live demo, no signup — paid plans open at launch
Article Extraction API
Any article URL in, clean structured text out — title, word count and main content, no nav bars or cookie walls.
Capabilities
- Strips navigation, ads, cookie walls and boilerplate automatically.
- Returns title, word count and the main content as text or Markdown.
- Works on any public article, blog post or documentation page.
- One call, structured JSON — no scraper or headless browser to maintain.
Where it fits
- Feed clean article text into an LLM pipeline or summarizer.
- Build a read-it-later app or custom RSS reader.
- Index external content for search without shipping HTML noise.
What comes out
IN · a noisy page
Article title · 1,240 wordsClean main content, no navigation, no ads, no cookie banners…
OUT · clean text
Try it live
try:
Public pages only · custom URLs: 5 extractions/day per visitor · nothing is stored
Call it from your code
curl -X GET "https://api.crateapi.com/extract/article?url=https://example.com/post" \ -H "X-API-Key: YOUR_KEY"
Trust & data handling
Every demo on this page runs against the real API — nothing is faked. Uploaded files and pasted text are processed in memory and never written to disk; nothing about your request is stored. Details: Trust & Security.
Pricing
This demo is free to use today. Document Tools plans open at launch — join the list and lock in the launch discount.
FAQ
Does it work on any site?
On public articles and blog posts generally, yes — heavily scripted or paywalled pages can vary.
Text or Markdown?
Both are available — pick the format that fits your pipeline.
Is the page stored?
No — it's fetched, parsed and returned; nothing is kept afterwards.
Ready to try it?
click anywhere or press Esc to close