Article Extractor — Clean Web Content to Markdown / Text

Extract the main article from any web page — as clean Markdown or plain text. This actor strips away nav bars, ads, cookie banners, sidebars and boilerplate, and returns just the content that matters, plus metadata: title, author, publish date, description, site name and language. Perfect for feeding clean text to LLMs, building datasets, or archiving content.

AI-agent ready: this Actor is callable and payable by AI agents via x402 (USDC on Base), no Apify account required.

Give it a URL, get back a clean article. No readability tuning, no HTML soup, no setup.

Who uses this

• 🤖 AI / LLM & RAG pipelines — get clean Markdown to embed, summarize or fine-tune on.
• 📰 Content & news monitoring — archive articles as readable text with metadata.
• 🔎 Researchers & analysts — build clean corpora from the open web.
• ✍️ Marketers & writers — pull competitor content and references without the clutter.
• 🧑‍💻 Developers — a reliable readability/boilerplate-removal endpoint by URL.

What you get — one clean row per URL

Field	Description
content	The main article as clean Markdown or plain text
title	Article title
author	Author / byline
date	Publish date
description	Meta description / summary
sitename	Publisher / site name
language	Detected language
wordCount	Word count of the extracted body
bodyFormat	markdown or txt
url, scrapedAt	Source & provenance

Example

Input

{ "urls": ["https://en.wikipedia.org/wiki/Web_scraping"], "outputFormat": "markdown" }

Output (one item)

{
  "url": "https://en.wikipedia.org/wiki/Web_scraping",
  "title": "Web scraping",
  "author": null,
  "date": "2026-05-01",
  "sitename": "Wikipedia",
  "language": "en",
  "bodyFormat": "markdown",
  "content": "# Web scraping\n\nWeb scraping is the process of automatically extracting data...",
  "wordCount": 3421
}

Why this extractor

• ✅ Boilerplate gone — nav, ads, cookie banners, sidebars and footers removed.
• 📝 Markdown or plain text — pick the format your pipeline needs.
• 🏷️ Rich metadata — title, author, date, description, site name and language.
• 🤖 LLM-ready — clean Markdown is ideal for RAG, summarization and fine-tuning.
• ⚡ Fast & affordable — pay only per article extracted.

Input options

• urls (required) — one or more page URLs (https:// added if missing).
• outputFormat — markdown (default) or txt.

FAQ

Markdown or plain text? Both — set outputFormat to markdown or txt.

Does it remove ads and navigation? Yes — it isolates the main article and strips nav, ads, cookie banners and boilerplate.

Can I process many URLs in one run? Yes — pass a list; each row carries its own url and metadata.

Is it good for LLM / RAG use? Yes — clean Markdown with metadata is exactly what embedding and summarization pipelines need.

How fresh is the data? Live — it fetches each page in real time on every run.

Can I export to CSV/Excel/Google Sheets? Yes — the dataset exports to JSON, CSV, Excel and HTML, or pull it via the API.

What does it cost? Pay-per-article — you're only charged for articles actually extracted.

---

Tip: pair this with the RSS Feed Reader — discover URLs from feeds, then extract full clean article text for your dataset.