Tools / web.markdown
Web page to Markdown
web.markdown
Converts a public web page to clean Markdown with headings, links, lists, tables and code kept.
WebLaunch set1 unit per call15 min cacheMCP: web_markdownv1.0.0
When to use it
Use to give a language model the content of a page with its structure: documentation, articles, READMEs. mode article keeps the main content only; mode page converts the whole page body without navigation. For plain text use web.extract.
Input
| Field | Type | Description |
|---|---|---|
url * | string (≤ 2048 chars) | Absolute http or https URL of the page. |
mode | one of article, page | article: main content only (falls back to page). page: the whole body without nav, header and footer. Default: "article". |
max_chars | integer (≥ 500, ≤ 200000) | Maximum characters of Markdown to return. Default: 50000. |
* required
Output
Returns data with: url, title, description, mode, markdown, chars, truncated.
Request
curl -X POST https://agentops.tools/v1/web.markdown \
-H 'content-type: application/json' \
-d '{"url":"https://en.wikipedia.org/wiki/Markdown"}'Over MCP the tool is named web_markdown. Over A2A its skill id is web.markdown. Same input, same envelope.
Try it
Counts against your rate limit.
Sources
Content from the source page; copyright remains with the publisher