How to Convert Any Web Page to Clean Markdown for an LLM

Follow the steps below and use the linked tools and references as you go.

Evidence trail: Web to Markdown API on RapidAPI · Web Content & URL Tools MCP Server on Apify

Feeding raw HTML to a language model wastes tokens on menus, scripts and footers. One GET request returns the page's main content as Markdown, with its title, word count and language.

Subscribe to the free plan on RapidAPI, then call the endpoint with your RapidAPI key:

curl -G 'https://web-to-markdown2.p.rapidapi.com/v1/markdown' 
  --data-urlencode 'url=https://avaloncompany.ai/' 
  -H 'x-rapidapi-host: web-to-markdown2.p.rapidapi.com' 
  -H 'x-rapidapi-key: YOUR_RAPIDAPI_KEY'

We ran the same check against our own site while writing this post. The key fields of the real response:

{
  "title": "Avalon — Make money with AI, learned from real experiments",
  "word_count": 352,
  "language": "en-US",
  "markdown_starts_with": "### Run a blog with AInnTopic research, drafting, editing and posting on a sched"
}

import requests

resp = requests.get(
    'https://web-to-markdown2.p.rapidapi.com/v1/markdown',
    params={'url': 'https://avaloncompany.ai/'},
    headers={'x-rapidapi-host': 'web-to-markdown2.p.rapidapi.com', 'x-rapidapi-key': 'YOUR_RAPIDAPI_KEY'},
    timeout=30,
)
resp.raise_for_status()
print(resp.json())

• Give an AI agent or RAG pipeline the readable text of an article or doc page.
• Summarise or translate a page without copying it by hand.
• Archive a page's content in a format that diffs cleanly.

• It reads the static HTML and does not run JavaScript, so pages that render their text in the browser come back short (the response includes a warning when that happens).
• Sites that block bots return an error instead of content.

The same endpoint is one of the tools in the Web Content & URL Tools MCP Server on Apify. Point any client that supports remote MCP servers at `https://avalonai–web-content-url-tools-mcp.apify.actor/mcp` with your Apify token (`YOUR_APIFY_TOKEN`) and the agent can call it directly: https://apify.com/avalonai/web-content-url-tools-mcp

Web to Markdown API on RapidAPI (free plan available): https://rapidapi.com/avalonai/api/web-to-markdown2

Our APIs on RapidAPI: https://rapidapi.com/user/avalonai

• Web to Markdown API — produced the example response in this tutorial (run against avaloncompany.ai): https://rapidapi.com/avalonai/api/web-to-markdown2