Web to Markdown — LLM Web Page Reader

Extract clean, LLM-ready Markdown and structured metadata from any webpage in milliseconds. Strip ads, popups, cookie notices, and navigation boilerplate so your AI agents, RAG pipelines, or fine-tuning datasets get pure content.

---

What does Web to Markdown Reader do?

Web to Markdown takes any public webpage URL and returns:

• Clean Markdown content: Headings, lists, code blocks, tables, and clean text.
• Page Metadata: Title, meta description, and canonical URL.
• Token Estimation: Pre-computed character and token counts (~len/4) for LLM context planning.
• Fast Execution: Uses lightweight extraction without heavy headless browser overhead.

Why use Web to Markdown Reader?

• Zero Bloat for LLMs: Raw HTML easily wastes 80-90% of token budgets on CSS, JS, and HTML boilerplate. This Actor filters straight to the article or main content.
• Pay Per Event (PPE): Pay only for what you extract, avoiding expensive monthly subscriptions.
• Agent & MCP Friendly: Easily integrated via Apify API, Apify MCP, or webhook automation.

---

How to use

1. Enter the Page URL you want to extract.
2. Toggle whether to include markdown hyperlinks or images.
3. Click Start or call via API.
4. Download or consume results directly in JSON, CSV, or Markdown.

Input Example

{
  "url": "https://news.ycombinator.com",
  "include_links": true,
  "include_images": false
}

Output Example

{
  "url": "https://news.ycombinator.com",
  "title": "Hacker News",
  "description": "",
  "content": "# Hacker News\n\n1. Qwen-Image-2.1: Compact, efficient...\n2. Key symbols we lost to time...",
  "length": 4623,
  "estimated_tokens": 1155,
  "elapsed_ms": 320
}

---

Output Fields

Field	Type	Description
url	string	Target webpage URL
title	string	Extracted webpage title
description	string	Meta description tag
content	string	Clean Markdown formatted text
length	integer	Character length of content
estimated_tokens	integer	Estimated LLM tokens (~length / 4)
elapsed_ms	integer	Execution time in milliseconds

---

API Integration

from apify_client import ApifyClient

client = ApifyClient("YOUR_APIFY_TOKEN")
run = client.actor("drain54/web-to-markdown").call(
    run_input={"url": "https://example.com"}
)

for item in client.dataset(run["defaultDatasetId"]).iterate_items():
    print(item["content"])