Tools / web.extract
Article text
web.extract
Main text, title, author and publish date from a public web page, as clean plain text.
WebLaunch set1 unit per call15 min cacheMCP: web_extractv1.0.0
When to use it
Use to read the content of an article, blog post or documentation page. Returns plain text, not HTML. Pages that need JavaScript to render may return EMPTY_CONTENT; web.screenshot can show those.
Input
| Field | Type | Description |
|---|---|---|
url * | string (≤ 2048 chars) | Absolute http or https URL of the page. |
max_chars | integer (≥ 500, ≤ 100000) | Maximum characters of text to return. Default: 20000. |
* required
Output
Returns data with: url, title, author, published, site_name, language, excerpt, text, word_count, truncated.
Request
curl -X POST https://agentops.tools/v1/web.extract \
-H 'content-type: application/json' \
-d '{"url":"https://en.wikipedia.org/wiki/Web_scraping"}'Over MCP the tool is named web_extract. Over A2A its skill id is web.extract. Same input, same envelope.
Try it
Counts against your rate limit.
Sources
Article text from the source page; copyright remains with the publisher