extract_page
One web page as typed JSON: an article, product, discussion or listing with its fields recognised automatically. ₹0.22 per page.
What it does
One web page read into typed fields, from url: the page type is recognised (article, product, discussion, image, video, list, event) and data.objects[] carries the fields for it: for an article the title, author, date, text, tags, sentiment and images; for a product the name, price, availability, brand and specs; for a discussion the posts. mode forces a type, fields adds links, meta or breadcrumb, discussion: false skips comments. Use it when you want fields rather than Markdown; for one page's text read_page, for several read_pages.
India: no locale knob; the page is read as served.
Inputs
Generated from the contract captured from staging on 2026-11-05.
| Field | Type | Required | Description |
|---|---|---|---|
url | string | yes | target URL to extract, URL-encoded. e.g. "https://www.technologyreview.com/2020/09/04/1008156/knowledge-graph-ai-reads-web-machine-learning-natural-language-processing/" |
mode | string | no | restrict extraction to one page type: article, product, discussion, image, video, list, event |
fields | string | no | comma-delimited optional fields: links, extlinks, meta, querystring, breadcrumb |
scroll | string | no | 'slow' or 'fast' — triggers lazy-loaded content |
timeout | integer | no | ms to wait for content retrieval. e.g. 30000 |
fallback | string | no | which API to fall back to if page type cannot be determined |
proxyAuth | string | no | username:password for the custom proxy |
discussion | boolean | no | pass false to disable automatic comment/review extraction. e.g. true |
renderDelay | integer | no | extra rendering ms before DOM extraction, max 180000 |
What drives the cost
Nothing: flat per page. Usually two to five seconds, up to twenty-five on a slow site.
Price today
| Price today | ₹0.22 per page (cost_paise: 22) at ₹93.93 per US dollar on 17 November 2026 |
| Hold before the call | ₹0.38 (hold_paise: 38); when the call settles the real price is charged and the unused part is released |
| ₹0 when | it fails, or the provider answers but there is nothing in it (an empty list, no email, no page): class: "ok_free", ₹0 |
| Timeout | 71 s, then upstream_timeout (₹0, retryable) |
| Over MCP | run_tool with tool: "extract_page" |
| Over REST | POST https://api.parel.si/v1/run with "tool": "extract_page" |
| Confirmation | not needed (the estimate is under the ₹200 threshold) |
Request
curl https://api.parel.si/v1/run \
-H "Authorization: Bearer $PAREL_KEY" \
-H "Content-Type: application/json" \
-d '{"tool":"extract_page","input":{"url":"https://en.wikipedia.org/wiki/Zomato"}}'
What comes back
data.objects[] and data.type. A real answer recorded during the live check of 2026-10-28, trimmed (long text cut):
{
"request": {
"options": [
"exclude=author",
"norender=true",
"… 1 more item(s) truncated"
],
"pageUrl": "https://en.wikipedia.org/wiki/Zomato",
"api": "analyze",
"version": 3
},
"humanLanguage": "en",
"objects": [
{
"date": "Wed, 07 Oct 2026 20:26:00 GMT",
"sentiment": 0.422,
"images": [
{
"naturalHeight": 0,
"width": 250,
"url": "https://thumb.wikimedia.org/wikipedia/commons/thumb/7/75/Zomato_logo.png/500px-Zomato_logo.png?utm_source=en.wikipedia.org&utm_campaign=parser&utm_content=thumbnail",
"naturalWidth": 0,
"primary": true,
"height": 250
},
{
"naturalHeight": 0,
"width": 148,
"title": "Zomato logo in 2012 ",
"url": "https://upload.wikimedia.org/wikipedia/commons/f/fd/Zomato-logo.png?utm_source=en.wikipedia.org&utm_campaign=parser&utm_content=thumbnail_unscaled",
"naturalWidth": 0,
"height": 51
},
"… 2 more item(s) truncated"
],
"shareLinks": false,
"icon": "https://en.wikipedia.org/static/apple-touch/wikipedia.png",
"siteName": "Wikimedia Foundation, Inc.",
"title": "Zomato",
"type": "article",
"tags": [
{
"score": 0.9545215368270874,
"sentiment": 0.941,
"count": 32,
"label": "Zomato",
"rdfTypes": [
"http://dbpedia.org/ontology/Organisation",
"http://dbpedia.org/ontology/Skill",
"… 1 more item(s) truncated"
]
},
{
"score": 0.8664276599884033,
"sentiment": 0.554,
"count": 4,
"label": "Deepinder Goyal",
"rdfTypes": [
"http://dbpedia.org/ontology/Person"
]
},
"… 3 more item(s) truncated"
],
"humanLanguage": "en",
"pageUrl": "https://en.wikipedia.org/wiki/Zomato",
"html": "<p>This article is about the food delivery service. For its owner, formerly known as Zomato Limited, see <a href=\"https://en.wikipedia.org/wiki/Eternal_Limited\">Eternal Limited</a>.</p>\n<table>Zomato<tbody><tr><td colspan=\"2\"><figure><a href=\"https://en.wikipedia.org/wiki/File:Zomato_logo.png\"><img data-file-height=\"1200\" data-file-type=\"bitmap\" data-file-wi…",
"categories": [
{
"score": 0.728,
"name": "Shopping",
"id": "iabv2-473"
},
{
"score": 0.728,
"name": "Grocery Shopping",
"id": "iabv2-477"
},
"… 4 more item(s) truncated"
],
"text": "Indian food delivery service\nThis article is about the food delivery service. For its owner, formerly known as Zomato Limited, see Eternal Limited.\nZomato\nFormer name\nFoodiebay (2008–2018)\nWebsite type\nOnline food ordering\nFounded\n10 July 2008; 18 years ago (2008-07-10)[1]\nHeadquarters\nGurugram, Haryana, India\nArea served\nIndia\nOwner\nEternal Limited\nFounders…"
}
],
"type": "article",
"title": "Zomato - Wikipedia"
}
Notes
- Flat per page. mode article or product when you know the type; the automatic choice is usually right.
- Long pages produce long text; over MCP long strings are shortened.
- Usually two to five seconds, up to twenty-five on a slow site.
Siblings
read_pagefor the page as Markdown.get_amazon_productfor an Amazon listing specifically.
This page as Markdown: /docs/tools/extract_page.md. Something wrong or missing: hello@parel.si.