Skip to main content

​
Base URL

A site’s own data, like a Reddit search or a YouTube transcript, comes from scrape with a parser. A field an endpoint does not take returns 400.

​
Authentication

Create a key at stophy.dev/signup and send it as a bearer token.
With no key, the endpoints that need one return 401 with code apiKeyRequired, a signupUrl for a person and a signup call for an agent. Search, scrape, map and parse also work without a key, within the free limits.

​
Responses

Each endpoint answers in its own shape, and each endpoint’s page shows a real response. Search returns success, data and creditsUsed:
Scrape and parse put the page in data, and scrape puts its cost in data.metadata.creditsUsed. Map returns links. Crawl, batch scrape and extract return a job id, and their status calls return status, the pages or fields in data, and creditsUsed. The account calls add requestId. Empty values are left out. Every response has an X-Request-ID header and an x-credits-used header. To cut long text, add ?maxTextLength=<characters> to the URL: longer text ends with …. A failed call returns an error status and a flat body: { "success": false, "error": "<what went wrong>", "code": "<code>" }. See Errors.

​
Cost

creditsUsed is what the call cost in credits. Most calls cost 1 credit, and a call that fails costs nothing. See Pricing.