ScrapingBee

ScrapingBee is a single-endpoint web scraping API: a client sends a GET request with a target URL and query parameters, and ScrapingBee returns rendered HTML, extracted data, or a screenshot, billing per successful request in "credits" that scale with the proxy tier and rendering options used. It also exposes several dedicated APIs for specific sources (Amazon, Google, Fast Search, Walmart, YouTube, ChatGPT, Gemini), a command-line client, and a hosted remote MCP server so AI agents can call the same capabilities as tools.

• Automatic proxy rotation with premium and stealth proxy tiers, plus country-level geotargeting
• AI-powered data extraction returning structured JSON or Markdown from a page
• CSS/XPath selector-based extraction rules
• Screenshot capture of rendered pages
• Auto-Mode that tries rendering/proxy configurations from cheapest to most expensive and charges only for the one that succeeds
• Dedicated pre-built APIs for Amazon, Google, Fast Search, YouTube, Walmart, ChatGPT, and Gemini
• Remote MCP server exposing scraping, search, e-commerce, video, and account-usage tools to AI assistants
• CLI with agent skills, no-code integrations (n8n, Make, Zapier) and SDKs (Python, Node.js, LangChain)

Developers, founders, and product teams that need web data for applications, internal tools, or AI-agent workflows and want to avoid running their own headless-browser and proxy infrastructure - named use cases include AI, e-commerce, market intelligence, GTM, fintech, and cybersecurity.

Teams scraping protected sites at volume should model those multipliers against plan allowances. This research did not retrieve a self-serve OpenAPI specification or interactive explorer, so teams that require a machine-readable contract should verify current availability with the vendor.

Setup assessment: 5m

Getting an API key and making a first successful scrape request is a copy-paste exercise from the quickstart, with no OAuth flow or account approval step; the free 1,000-credit tier requires no credit card.

Prerequisites

• ScrapingBee account and API key

Authentication

API key, sent as an Authorization: Bearer header (recommended) or as an api_key query parameter (deprecated but supported)

Base URL

https://app.scrapingbee.com/api/v1

Machine-readable API

No OpenAPI link in the reviewed source set

• Single GET endpoint (HTML API) accepting a target URL plus parameters controlling JS rendering, proxy type, geotargeting, extraction rules, AI extraction, screenshots, and Auto-Mode (mode=auto)
• Dedicated Google, Fast Search, Amazon, Walmart, Shopee, ChatGPT, Gemini, YouTube, and Employee Search APIs
• /usage endpoint for real-time credit and concurrency usage (6 calls per minute)

• No official YouTube channel link was found in the site footer or documentation, so no officialVideo is recorded
• SLA terms and the SOC 2 Type II report contents (report is gated behind a Trust Center access request)
• A machine-readable OpenAPI specification was not included in the reviewed source set.