Chocodata

Features

Structured JSON from 237 sites across 17 categories - search engines, e-commerce, social, real estate, finance and more - via a single REST endpoint. Built for developers.

Get Started View Documentation

What you can extract

237 targets, one consistent request/response shape. The product and search examples below show the JSON every endpoint returns.

Core endpoints

shopping_bag

Product data

Dozens of structured fields per product page: title, price, strikethrough, variations, seller offers, ratings, top reviews, category ladder, images, bullets.

search

Search results

Keyword-driven SERP data - organic positions, sponsored ads, prices, ratings, and image URLs for every product on the page.

Coming soon reviews

Paginated reviews

Full-review extraction with pagination and star/recency filters. Current product endpoint includes the top 8–10 reviews; deep pagination is on the roadmap.

Infrastructure

vpn_lock

Residential proxies

Every request egresses from a residential IP in the matching country automatically - a German site via a DE IP, a Japanese site via a JP IP. No configuration needed.

shield

Anti-bot handling

WAF and CAPTCHA challenge detection + automatic retry escalation. Most scrapes succeed on first try; the rest fall through a 2-pass retry loop transparent to you.

language

237 targets across 17 categories

Search engines, e-commerce, social, real estate, finance, knowledge bases and more - one endpoint for the whole web. Geo-target by country and override the content language per request where the target supports it.

Advanced

Coming soon code

JavaScript rendering

Full headless-Chromium render for pages whose data requires client-side JS. On the roadmap - the `render_js` parameter currently returns 501.

batch_prediction

Async batch + webhooks

POST up to 1,000 product IDs, URLs, or search queries in a single request. Poll the batch status endpoint or receive results via signed webhook callback when processing completes.

data_object

Structured JSON output

Typed, predictable response shape. Validated, parity-checked across endpoints. Ready for direct ingestion into databases, analytics, and ML pipelines.

Drop-in integrations

First-class SDKs for the three languages buyers of scraping infrastructure actually ship to production, plus a CLI for one-offs and an MCP server so AI agents can pull web data natively.

javascript

npm

Node.js SDK

arrow_forward 
npm install chocodata

JavaScript / TypeScript

PyPI

Python SDK

arrow_forward 
pip install chocodata

Python 3.9+

Go modules

Go SDK

arrow_forward 
go get github.com/Chocodata-com/chocodata-go

Go 1.21+
terminal

Homebrew / curl

CLI

arrow_forward 
brew install chocodata-com/tap/chocodata

One-off + scripted runs
smart_toy

Model Context Protocol

MCP server

arrow_forward 
npx -y @chocodata/mcp-server

Claude / Cursor / MCP host

Prefer raw HTTP? Any language that speaks JSON over HTTP works - copy-paste snippets for cURL, Ruby, PHP, Java, and C# live in the dashboard playground.

Shipping vs. roadmap

Every plan - including the free tier - gets every shipping feature and all 237 targets. Tiers only change volume, concurrency, and support.

Live in production

  • check_circleProduct data (dozens of structured fields)
  • check_circleSearch results with organic/sponsored positions
  • check_circle237 targets across 17 categories (250+ with validated JSON, 230+ dedicated specific-item endpoints)
  • check_circleGeo-targeting + content-language override where supported
  • check_circleResidential proxies, country-matched automatically
  • check_circleAWS WAF challenge detection + 2-pass retry
  • check_circleTop reviews embedded in product response
  • check_circleFull response HTML on demand (`add_html`)
  • check_circleOnly-2xx billing - errors don't cost credits
  • check_circlePer-key usage + analytics in the dashboard
  • check_circleAsync batch + signed-webhook delivery (up to 1,000 items per call)
  • check_circleOfficial SDKs for Node, Python, Go, + CLI + MCP server

On the roadmap

  • schedulePaginated reviews endpoint - deep review extraction past the product page's top-10.
  • scheduleJavaScript rendering - `render_js=true` for JS-heavy pages.
  • scheduleFull-page screenshots - PNG capture of the rendered page.
  • scheduleDedicated sub-endpoints - best-sellers, Q&A, lightning deals.

Ready to scrape the whole web?

Free tier: 1,000 requests / month across all 237 targets. No credit card required.

Get Started Read the Docs