Docs

Documentation

Chocodata is one scraping API for the whole web - a single REST request returns structured JSON from 237 sites across 470 endpoints. Everything below covers making your first scrape, shipping a production integration, and how billing works. The endpoints below illustrate the shared request/response shape that every target follows; browse the full catalogue in the scraper directory. For a hands-on path, start with Getting started.

Getting started

Make your first scrape in under 2 minutes - get a key, run one cURL, parse the response.

Core concepts

How the API works: the one request shape, dedicated endpoints vs the Universal scraper, the credit model.

Endpoint reference

How 470 endpoints share one shape, the resource types, and how to find any endpoint in the directory.

Authentication

API key format, rotation, and hygiene. Unauthorized requests never cost credits.

Product endpoint

Single item-page scraping - dozens of structured fields including price, variations, seller offers.

Search endpoint

Ranked search results across search engines, marketplaces, and boards: positions, prices, ratings, images.

Universal Web Scraper API

Scrape any URL to JSON, clean HTML, or text - for sites without a dedicated endpoint.

Batch endpoint (async)

Submit up to 1,000 IDs, URLs, or queries in one call. Webhook or polled results.

Error codes

Complete list of error codes with retry semantics. Only 2xx costs credits, always.

Rate limits & concurrency

Per-plan ceilings on RPS and concurrent in-flight requests. How to check utilisation and request more.

SDK quick-starts

Install + first-request snippets for the official Node, Python, Go, and CLI SDKs.

Country, region & language

How we route to the right regional site and which languages each one actually serves.

Billing policy

Which responses cost credits - simple rule: only successful (2xx) responses are billed.

Changelog

Every visible change to the API, SDKs, and dashboard. Breaking changes get 14 days of notice.

Looking for an SDK?

We publish official SDKs for Node (chocodata), Python (chocodata), Go (github.com/Chocodata-com/chocodata-go), a cross-platform CLI (chocodata-cli), and an MCP server (chocodata-mcp) for AI agents. Each ships with the same auth + retry behaviour. See SDKs, CLI & MCP.

For coding agents & LLMs

The entire docs corpus is available in LLM-friendly formats:

· /docs/llms.txt - manifest of every page with URLs and descriptions (llms.txt spec).
· /docs/llms-full.txt - single-file concatenation of every page for one-shot ingestion.
· /docs/<slug>.md - append .md to any doc URL for the raw Markdown source. Example: /docs/endpoints/product.md.

Every page also carries "Copy as Markdown", "Open in ChatGPT", and "Open in Claude" buttons in the top-right so you can hand the page to an agent in one click.