Website to Markdown Scraper - HTML to MD for RAG API

Convert web pages and HTML to clean Markdown for RAG, LLM training, AI knowledge bases, and content migration. Strips nav, ads, scripts, and styling while preserving structure. Bulk output includes word/link/image counts.

Website to Markdown Converter for LLM Training | $0.002/page

Convert any web page to clean, structured Markdown optimized for LLM training, RAG pipelines, and content processing. Process hundreds of URLs in a single run with concurrent fetching.

What it does

• Fetches web pages and extracts the main readable content
• Strips navigation, ads, footers, scripts, styles, and other non-content elements
• Converts HTML to Markdown preserving structure: headings, paragraphs, lists, tables, code blocks, blockquotes, and inline formatting
• Optionally includes links and images as Markdown references
• Reports metadata per page: word count, character count, link count, image count

Use cases

• LLM training data -- Convert web content to clean text for fine-tuning datasets and RAG retrieval pipelines
• AI knowledge bases -- Build structured Markdown knowledge bases from documentation sites, wikis, and help centers
• Content migration -- Move content between CMS platforms (WordPress, Ghost, Notion) in portable Markdown format
• Documentation archival -- Archive web-based docs as version-controlled Markdown files
• SEO content auditing -- Extract and analyze competitor content structure and word counts at scale
• Text analysis -- Clean text extraction for NLP processing, sentiment analysis, and topic modeling
• Research data collection -- Scrape articles, blog posts, and papers into a structured dataset

Input

Field	Type	Default	Description
urls	string[]	required	List of URLs to convert
maxResults	integer	100	Maximum pages to process
includeLinks	boolean	true	Keep hyperlinks in the Markdown output
includeImages	boolean	false	Keep image references in the Markdown output

Output fields

Each result in the dataset includes:

Field	Description
url	Source page URL
title	Page title
markdown	Full Markdown content with structure preserved
wordCount	Number of words in the extracted content
characterCount	Total characters in the Markdown output
linkCount	Number of hyperlinks found
imageCount	Number of images found

Example output

{
  "url": "https://en.wikipedia.org/wiki/Web_scraping",
  "title": "Web scraping - Wikipedia",
  "markdown": "## Web scraping\n\nWeb scraping is data scraping used for extracting data from websites...",
  "wordCount": 3847,
  "characterCount": 24102,
  "linkCount": 156,
  "imageCount": 2
}

Pricing

$0.002 per page converted. Convert 500 pages for $1.

Integrations

Works with Apify's API, webhooks, and MCP server integration. Chain with other actors to build automated content pipelines.

How to use

1. Open Website to Markdown Scraper - HTML to MD for RAG API in Apify Console and click Start (or Try for free).
2. Fill in the input form - start with Page URLs, Max Results, Include Links. Every field is optional unless marked required, and the form is prefilled with a working example.
3. Set Maximum results to cap your cost, then run the Actor.
4. Download the results as JSON, CSV, Excel, XML or HTML, or pull them with the Apify API.

Example input

{
  "urls": [
    "https://en.wikipedia.org/wiki/Web_scraping",
    "https://docs.apify.com/academy/getting-started"
  ],
  "maxResults": 2,
  "includeLinks": true,
  "includeImages": false
}

FAQ

How much does Website to Markdown Scraper - HTML to MD for RAG API cost?

$0.002 per result plus $0.0001 per run. For example, 1,000 results cost about $2.00. Higher Apify plans get a Store discount.

Do I need an API key or login?

No. Just configure the input and run it.

Can I run it on a schedule or from code?

Yes. Use Apify Schedules for recurring runs, or call it from the Apify API, the JavaScript or Python client, Make, Zapier, n8n, or any MCP-compatible AI agent.

Why did my run return no results?

Your filters are probably too narrow. Remove a filter or broaden the search, then run again. You only pay the result price for records that are actually saved.

Is the data accurate and up to date?

Each run fetches live data from the source described above, so results are as current as that source. Verify critical records with the original source before acting on them.