Scrape web page
Fetch a web page and return readable text plus requested markdown, HTML, links, metadata, or schema-guided JSON.
POST /api/v1/platform/scrapers/web/scrape
Scrape a single web page and return its content.Scrape a single web page and return readable content. By default the endpoint returns the backward-compatible
url, text, and statusCode fields. Use
formats when you need markdown, HTML, discovered links, page metadata, or
schema-guided JSON extraction.
Example
Formats
formats is optional and defaults to ["text"].
data. If a
requested format is unavailable, the field may be null or an empty array.
Schema-guided JSON
Requestjson only when you provide jsonSchema. The schema defines the
returned keys and shape; page content is mapped into that schema and validated
before the response is returned. The root schema must be an object; use array
properties for repeated data such as comments, products, or plans.
jsonSchema and instructions are only valid when formats includes json.
Schema-guided extraction performs additional model work and may use more credits
than a plain text or markdown scrape.Authorizations
The public API key from the Credentials page.
The API secret shown when the credential is created.
Body
URL to scrape
"https://example.com/page"
Render JavaScript before scraping
Output formats to return. Defaults to ["text"]. Request "json" only with jsonSchema.
text, markdown, html, links, metadata, json JSON Schema used for schema-guided extraction when formats includes "json".
Additional extraction instructions used only when formats includes "json".
8000Maximum accepted cache age in milliseconds. Use 0 or omit to force a fresh scrape.
0 <= x <= 2592000000Maximum accepted cache age in days. Use 0 or omit to force a fresh scrape. Ignored when maxAgeMs is provided.
0 <= x <= 30