Website Content Crawler for AI & RAG: Markdown — $0.40 per 1,000 pages

Turn any website into clean text and Markdown ready for LLMs, RAG pipelines, vector databases, chatbots and search. Give it a start URL; it follows links on the same site, strips menus, footers, cookie banners and ads, and returns each page's main content with headings, lists, links, code blocks and tables kept.

It uses the same input and output fields as Apify's own Website Content Crawler (startUrls, maxCrawlPages, maxCrawlDepth, includeUrlGlobs, excludeUrlGlobs, removeElementsCssSelector; output url, crawl, metadata, text, markdown), so existing integrations and prompts keep working.

Why use it

• Low price: $0.40 per 1,000 pages, a flat price you can see before you run.
• Fast: it reads server HTML directly (no browser), so thousands of pages take minutes.
• Clean output: main-content detection, Markdown with structure, per-page title, description, language and Open Graph data.
• Polite: respects robots.txt, adjustable concurrency, same-site only.
• Control: include and exclude URL patterns, depth and page limits, sitemap discovery.

Pricing

Event	Price
Page crawled	$0.40 per 1,000 ($0.0004 each)
Actor start	$0.00005

Free Apify accounts are limited to 25 pages per run.

Input

{
  "startUrls": [{ "url": "https://docs.example.com" }],
  "maxCrawlPages": 500,
  "maxCrawlDepth": 5,
  "includeUrlGlobs": [{ "glob": "https://docs.example.com/guides/**" }],
  "excludeUrlGlobs": [{ "glob": "https://docs.example.com/**/changelog/**" }],
  "useSitemaps": true,
  "saveMarkdown": true
}

Output

{
  "url": "https://docs.example.com/guides/start",
  "crawl": { "loadedUrl": "https://docs.example.com/guides/start", "loadedTime": "2026-10-19T10:00:00.000Z", "referrerUrl": "https://docs.example.com/", "depth": 1, "httpStatusCode": 200 },
  "metadata": { "canonicalUrl": "https://docs.example.com/guides/start", "title": "Getting started", "description": "...", "languageCode": "en", "openGraph": { "title": "...", "image": "...", "type": "website", "siteName": "Example" } },
  "screenshotUrl": null,
  "text": "Getting started\nInstall the package...",
  "markdown": "# Getting started\n\nInstall the package...\n\n```\nnpm install example\n```"
}

Honest limits

• No JavaScript rendering. Pages that only build their content in the browser (some single-page apps) return little or nothing. Sites that serve their content in HTML, which includes most documentation sites, blogs, news and company sites, work well.
• No screenshots (screenshotUrl is always null) and no login or form handling.
• Some sites block automated requests. Turn on Apify Proxy in the input to retry through a proxy.
• Content extraction is automatic: unusual page layouts can include or miss some text. Use "Remove elements" to tune it.

Feedback and reviews

Found a bug, or a page that came out badly? Open an issue on this Actor's Issues tab with the URL and it will be looked at quickly. If this Actor saved you time, a short and honest review helps other people find it. Thank you!

Legal

Only crawl content you are allowed to use. The crawler respects robots.txt by default; respect each site's terms and copyright when you reuse the content.