Web page to Markdown
Extract readable Markdown, text and links from a public page.
This check runs on the server. Input and results are not saved as execution history; hosted allowances apply.
About Web page to Markdown
Fetch a public web page and extract readable Markdown, plain text, headings, links and page metadata.
When to use it
Give an agent a clean document instead of a whole HTML page.
A documentation URL → title, Markdown, text and absolute linksHow to use it
- Enter a public HTTP or HTTPS URL.
- Extract the page without executing JavaScript.
- Copy the Markdown or inspect its headings and links.
A readable source for agents
The extractor follows public redirects, then looks for a main or article element before falling back to the page body. It removes navigation, forms, scripts and images, keeps headings, lists and code blocks, and resolves page links to absolute URLs. Title, description and canonical metadata stay alongside the content rather than being mixed into the Markdown.
Use the Markdown for an agent context or fixture and the isolated preview to check what was extracted. The plain text and link list are available separately. Pages that depend on JavaScript, authentication or a paywall may return little useful content. The extractor does not bypass those restrictions or download linked assets.
Limits & privacy
Public network protection on every redirect, ten seconds and 256 KiB response data, 5,000 elements and 64 nesting levels. Scripts, navigation and forms are removed. JavaScript rendered or paywalled content may be absent. Images and linked pages are not fetched. No cookies or credentials are sent.
Use this tool from the API reference or connect through the SDK, CLI, SSH, or MCP. Hosted calls share your plan allowances.