Convert a web page to Markdown

flatmark converts HTML that you send and does not fetch URLs. Download the page with your own client, then send the HTML to POST /v1/convert.

curl

curl -sL https://example.com/ -o article.html
curl -F "file=@article.html;type=text/html" https://api.flatmark.dev/v1/convert

Or in one pipe, with the HTML read from stdin:

curl -sL https://example.com/ | \
  curl -F "file=@-;type=text/html;filename=article.html" https://api.flatmark.dev/v1/convert

Python

import httpx

html = httpx.get("https://example.com/", follow_redirects=True).text
resp = httpx.post(
    "https://api.flatmark.dev/v1/convert",
    files={"file": ("article.html", html.encode(), "text/html")},
    timeout=60,
)
resp.raise_for_status()
print(resp.json()["markdown"])

Before you send

• Navigation, headers and footers stay in the output. Send only the main element if you want the article alone.
• A page that builds its content with JavaScript has little HTML to convert. Render it in a headless browser first and send the result.

Next step

Paste a page's HTML into the free tool and check the output before you automate it.

FAQ

Can flatmark fetch a URL for me?

No. flatmark converts files and HTML that you send. Fetch the page with your own client, headers and cookies, then send the HTML.

What about pages that load their content with JavaScript?

flatmark converts only the HTML you send and runs no JavaScript. Render the page in a headless browser first, then send the result.