Webpage to Markdown API

Convert a public webpage into Markdown for your documentation or content workflow. The API returns the converted text in data.result, ready to save, review, or split into smaller sections.

View API details or try the Demo.

What you can build

Import public documentation into a knowledge base.
Prepare articles for indexing, classification, or summarization.
Review webpage content in a text-based editorial workflow.
Keep Markdown copies alongside source URLs for later reference.

Markdown can retain useful structure such as headings, lists, and links, but results depend on the source page. Review representative pages before relying on a specific layout, table structure, or main-article-only extraction. This is not a guarantee that every page element or access-restricted page can be converted.

Request

Send POST https://api.gugudata.io/v1/websitetools/url2markdown. Place appkey in the query string and url in the JSON body.

Parameter	Location	Required	Description
`appkey`	Query	Yes	Your application key. Keep it in server-side configuration.
`url`	JSON body	Yes	Complete public HTTP or HTTPS URL, up to 2,048 characters.

URLs containing credentials, IP addresses, or non-standard ports are not accepted. Use the source URL including any query parameters needed to select its content.

curl --request POST \
  'https://api.gugudata.io/v1/websitetools/url2markdown?appkey=YOUR_APPKEY' \
  --header 'Content-Type: application/json' \
  --data '{"url":"https://gugudata.io/"}'

Response

A successful response contains one Markdown string. The following example is illustrative; the homepage text changes over time.

{
  "dataStatus": {
    "requestParameter": "url_host=gugudata.io&url_path=/",
    "statusCode": 200,
    "status": "SUCCESS",
    "statusDescription": "successfully",
    "responseDateTime": "2026-09-08 00:00:00+0000",
    "dataTotalCount": 1
  },
  "data": {
    "result": "# GuGuData\n\nExplore APIs for your applications.\n"
  }
}

Field	Type	Meaning
`data.result`	String	Non-empty Markdown text. Whitespace is preserved.
`dataStatus.statusCode`	Integer	Business status; `200` means conversion succeeded.
`dataStatus.requestParameter`	String	Source host and path, without query parameters or the AppKey.
`dataStatus.responseDateTime`	String	Response time in `yyyy-MM-dd HH:mm:ssZ` format.

Failed conversions do not return a substitute article or a successful empty string. Check the HTTP status and business status before using the result.

Node.js example

async function toMarkdown(sourceUrl) {
  const endpoint = new URL(
    "https://api.gugudata.io/v1/websitetools/url2markdown",
  );
  endpoint.searchParams.set("appkey", process.env.GUGUDATA_APPKEY);

  const response = await fetch(endpoint, {
    method: "POST",
    headers: { "Content-Type": "application/json" },
    body: JSON.stringify({ url: sourceUrl }),
  });
  if (!response.ok) {
    throw new Error(`Conversion failed: HTTP ${response.status}`);
  }
  const body = await response.json();
  if (body.dataStatus?.statusCode !== 200 ||
      typeof body.data?.result !== "string" ||
      !body.data.result.trim()) {
    throw new Error("No usable Markdown was returned");
  }
  return body.data.result;
}

Store the source URL and retrieval date with the Markdown when you need provenance. Avoid logging credentials or sensitive URL parameters. Treat converted content as untrusted input, including when rendering it as HTML or passing it to an AI model.

Demo and caching

The Demo converts the GuGuData homepage. Successful Demo results are cached for 15 minutes. The X-GuGuData-Demo-Mode header reports live or cached, and X-GuGuData-Generated-At records when the sample was generated. Failures are not cached. Use the authenticated endpoint for your own URLs.

Errors

HTTP status	Meaning
`400`	Missing or invalid URL, or invalid JSON request.
`401` / `403`	Missing, invalid, or unauthorized application key.
`429`	Request limit reached.
`500`	Unexpected service error.
`503`	Conversion could not be completed; business status is `500`.

Keep retries bounded and do not repeatedly submit unchanged invalid input. Blocked pages, unavailable sources, and unusually large results may fail. Do not interpret conversion success as proof that all source content was captured.