Wikipedia Scraper — Article Summaries, Images & URLs via the Official API

Launch pricing: this actor currently adds no fee — run it on your Apify free-plan credits. Try it, bookmark it, wire it into your stack.

A Wikipedia scraper on the official MediaWiki API: scrape Wikipedia articles by search term or exact titles and get the intro summary, page image, canonical URL and last-edited date as clean JSON. As a Wikipedia API scraper it's never blocked, works in every language edition, and doubles as a Wikipedia article extractor for Wikipedia data extraction into RAG and enrichment pipelines.

Canonical knowledge, structured: search → top articles → clean summaries, in any of 300+ language editions.

Who uses this

• RAG / LLM builders — intro extracts are dense, neutral grounding context with a canonical URL for citations.
• Data enrichment teams — resolve entity names to a description, image and stable URL in one call.
• Researchers & content teams — build topic corpora from any language edition.
• Monitoring — diff lastEdited across scheduled runs to catch article changes.

Grounding LLMs on live tech discussion too? The Hacker News Scraper delivers the same clean JSON for the HN front page.

How it works

The actor resolves your input to article titles — either a MediaWiki full-text search on searchTerm or your explicit titles list — then fetches intro extracts, page images and canonical URLs in batches of 20 (the API's extract limit), following redirects. Failed searches, all-batches-failed and zero-article runs fail loudly instead of succeeding empty.

Output

One record per article:

{
  "title": "Anthropic",
  "pageId": 6206236,
  "extract": "Anthropic, PBC is an American artificial intelligence (AI) company headquartered in San Francisco, California. It has developed a series of large language models (LLMs) named Claude and has a focus on AI safety...",
  "url": "https://en.wikipedia.org/wiki/Anthropic",
  "thumbnail": null,
  "lastEdited": "2026-07-15T05:56:25Z",
  "language": "en",
  "scrapedAt": "2026-07-15T09:35:19.852Z"
}

• extract is the article's intro section as plain text (no wiki markup, no HTML).
• thumbnail is the page image at 320px, or null when the article has none.
• url is the canonical article URL — use it for citations.

Input

Field	Default	Description
searchTerm	OpenAI	Keyword to search; top matches are scraped. Ignored when titles is set
titles	—	Exact article titles to fetch directly
language	en	Wikipedia language edition (en, de, fr, es, …)
limit	10	Max articles when searching by term

The default run needs no configuration — an empty input searches OpenAI on English Wikipedia.

Reliability & limits

• Extracts are intro sections only, not full article bodies — ideal for summaries/grounding, not for full-text mining.
• The MediaWiki API caps plain-text extracts at 20 per request; the actor batches automatically.
• Content freshness is live (straight from the API); lastEdited tells you exactly how fresh each article is.

Paste this output into…

• Google Sheets — =IMPORTDATA("https://api.apify.com/v2/acts/exuberant_volley~wikipedia-scraper/runs/last/dataset/items?format=csv&clean=1&token=YOUR_TOKEN") for an instant reference table.
• Make / n8n — feed a list of entity names as titles, fetch .../runs/last/dataset/items, and write extract + url back into your CRM or Airtable.
• RAG ingestion (LLM builders) — chunk extract, embed it, and store url + lastEdited as metadata — you get citable, dated grounding snippets with zero cleaning.

ScrapeForge free data suite

One publisher, ten plug-and-play datasets — all currently free to run:

Actor	What it delivers
Executive Changes Tracker	New CEO/CFO/board moves from SEC 8-Ks + official newswires — source-cited B2B trigger leads	run it next →
IKEA Product Scraper	Names, prices, ratings and images from any IKEA search or category page	run it next →
Shopify Store Scraper	Any Shopify store's full catalog — variants, SKUs, prices — via products.json	run it next →
App Store Apps Scraper	App rankings, ratings, prices and metadata by keyword, any country	run it next →
Remote Jobs Scraper	Live remote listings from RemoteOK + Remotive, keyword-filtered	run it next →
CoinGecko Market Scraper	Top-coin prices, market caps and 24h moves — keyless	run it next →
GitHub Repositories Scraper	Repo search with stars, forks, topics and licenses	run it next →
Hacker News Scraper	Top/New/Best HN stories with scores and comment counts	run it next →
Wikipedia Scraper	Article summaries, images and URLs in any language — RAG-ready	run it next →
SEC EDGAR Filings Scraper	Any US public company's filing history by ticker	run it next →

Compliance

• No personal data — public encyclopedia content only.
• Official API underneath — the MediaWiki API with a descriptive User-Agent; no HTML parsing.
• Wikipedia text is CC BY-SA — keep attribution (the url field) when you republish extracts.
• Records live only in your run's dataset; the actor keeps nothing beyond it.