Español

Lee este artículo en español.

Web Scraping in Make: Get Any Page as Text or Markdown

Short answer: The free way is HTTP → Make a request (a GET to the page) followed by Text parser → HTML to text, which gives you plain text. When you want Markdown that keeps headings, links and tables, only the main content of the page, or a page that needs JavaScript, the Cofferdock URL to Markdown API is the better fit.

Make can read a web page with its built-in HTTP module, and for many jobs that is enough. This guide shows the free way first, then explains when a dedicated API saves you work. Use the free way whenever it does the job for you.

The free way: HTTP module + HTML to text

  1. Add the HTTP module and choose Make a request.
  2. Set the URL to the page address, and the Method to GET.
  3. Run the scenario once. The HTML of the page now comes out of this module in a field called data.
  4. Add the Text parser module and choose HTML to text. In its HTML field, insert {{1.data}} (the 1 is the number of your HTTP module).
  5. The result is plain text. You can use it in any later module.

You can also skip the Text parser. In any field, write {{stripHTML(1.data)}} and Make removes the HTML tags for you.

Where the free way falls short

The API way: Cofferdock URL to Markdown

Use this when you want Markdown or a cleaner result. Add an HTTP module, choose Make a request, and set:

The page text comes back in the content field (for example {{2.content}}), and the page title in title.

A quick example

A Google Sheets sheet has one web address in each row. Make reads each row, sends the address to the API with the HTTP module, and then saves the content field into Notion or Google Docs. You can also send it to an AI module. If you do, remember that the text comes from someone else's website, so treat it as data and never as instructions.

Options you can tweak

You can add these options to the JSON body, next to url.

OptionDefaultWhat it does
formatmarkdownmarkdown, text (plain text) or metadata (only title, description, image...)
main_contenttrueKeeps only the main content. Set false to get the whole page
include_linkstrueKeeps the links. Set false to remove them
include_imagesfalseSet true to add images as Markdown links
max_chars200,000Longest text returned (100 to 1,000,000). Longer text is cut and truncated is true

Prefer to see it in code?

curl -X POST "https://api.cofferdock.com/web-to-markdown" \
  -H "x-api-key: YOUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{"url": "https://example.com/article"}'

Replace YOUR_KEY with your own key.

How much does it cost?

Each page read uses 1 credit, and failed calls are not charged. The free plan gives you 50 credits a month with no card. The same credits work for every Cofferdock tool.

FAQ

Can Make scrape for free? Yes. The HTTP module and the HTML to text module come built into Make, and they read simple pages without a Cofferdock account.

Why plain text and not Markdown? Make's built-in text tools return plain text. If you need headings, links and tables kept as Markdown, use the API.

Does it work on pages built with JavaScript? The free way does not run JavaScript, so those pages may come back almost empty. The Cofferdock API opens the page in a real browser with JavaScript turned on.

Is scraping allowed? Only read pages you are allowed to read, and respect each site's terms. The tool cannot read pages that need a login.

Related guides

Try it yourself

Try the tool on any article at cofferdock.com/web-to-markdown. You can start free at cofferdock.com/register, with no card needed.