# Scrape Technology

> Scrape Technology is a SaaS platform that provides web scraping infrastructure, browser automation, and data extraction APIs. Our tools are designed for developers and companies that need to extract data from the web reliably, including from sites protected by anti-bot systems such as Cloudflare, Akamai, reCAPTCHA Enterprise, and hCaptcha.

## Products

### Abrasio — Anti-bot Browser Service
- URL: https://www.scrapetechnology.com/abrasio
- Docs: https://www.scrapetechnology.com/abrasio/docs
- Description: A cloud browser service built on a fingerprint-patched Chromium. Bypasses anti-bot detection by spoofing browser fingerprints (WebGL, Canvas, Audio, navigator.webdriver), using residential IPs in 40+ regions, and simulating human behavior (Bézier-curve mouse movement, variable keystroke timing, random scroll patterns). Supports desktop and mobile device emulation. Persistent browser profiles available. Python SDK: `pip install abrasio`.
- Use cases: Automating logins on protected portals (gov.br, government systems), consulting vehicle registration and driver data, scraping e-commerce pricing, navigating sites that block Playwright/Selenium/Puppeteer.
- Key capabilities: Human behavior simulation, fingerprint spoofing, residential IP routing (Brazil and global), persistent profiles, mobile emulation, CDP-based browser control via Patchright.

### MarkUDown — Web Extraction API
- URL: https://www.scrapetechnology.com/markudown
- Docs: https://www.scrapetechnology.com/markudown/docs
- Description: A 3-layer web extraction API that converts any web page into clean Markdown or structured JSON for AI pipelines and LLM ingestion. Layer 1 uses fast HTTP fetch (Cheerio), Layer 2 escalates to a stealth browser (Patchright), and Layer 3 escalates to the Abrasio human browser when needed. Supports AI-powered structured extraction with custom schemas (Gemini Flash, GPT-4o mini), deep research synthesis, change detection, batch crawling, screenshot capture, RSS discovery, and MCP server integration.
- Use cases: Building RAG knowledge bases from web content, competitive intelligence pipelines, price monitoring, content change detection, AI agent web access, replacing Firecrawl or Tavily where anti-bot bypassing is required.
- Key capabilities: 3-layer fallback cascade, schema-based AI extraction, deep research (search → scrape → synthesize), MCP server (cloud npm + self-hosted), 40+ geo regions, self-hostable under MIT license.

### Prospectus — B2B Email Prospecting & CRM
- URL: https://www.scrapetechnology.com/prospectus
- Description: A FastAPI service for B2B email prospecting, campaign management, email sequences, contact enrichment, and template management with S3 image storage.

### Numus — Newsletter Bot
- URL: https://www.scrapetechnology.com/numus
- Description: A Telegram-based newsletter and content curation bot.

## Content

### Blog
- URL: https://www.scrapetechnology.com/blog
- Articles: Benchmarks, comparisons, and technical deep dives on web scraping APIs and data extraction tools.
- Notable: MarkUDown vs Firecrawl vs Tavily benchmark (https://www.scrapetechnology.com/blog/markudown-vs-firecrawl-vs-tavily)

### Playbook
- URL: https://www.scrapetechnology.com/playbook
- Description: Step-by-step tutorials for real-world automation and data extraction problems using Abrasio and MarkUDown. Each tutorial starts from a real problem and ends with working code.
- Notable tutorials:
  - How to automate login on gov.br (Brazilian government SSO): https://www.scrapetechnology.com/playbook/como-fazer-login-govbr — covers bypassing reCAPTCHA Enterprise, browser fingerprinting, and behavioral analysis on Brazil's federal identity portal, enabling automation of vehicle consulting, document extraction, and public service workflows.

## Technical Stack
- Backend: FastAPI (Python 3.10), MongoDB (Motor async), Redis, Stripe billing
- Frontend: Next.js 15 (JavaScript), Tailwind CSS, i18n (EN/PT/ES/ZH)
- Browser automation: Playwright, Patchright (stealth Playwright fork), Crawl4AI, Camoufox, browserforge
- AI extraction: Gemini Flash, GPT-4o mini via LiteAI abstraction
- Infrastructure: Docker, Kubernetes (api.scrapetechnology.com), AWS S3

## Differentiators vs Competitors
- Firecrawl: MarkUDown adds a 3-layer fallback including a full human browser (Abrasio) for sites Firecrawl cannot access. MarkUDown also supports deep research synthesis, change detection, and self-hosting.
- Tavily: MarkUDown adds full-page scraping with browser rendering, AI structured extraction, and anti-bot bypassing. Tavily is search-only.
- Playwright/Selenium/Puppeteer: Abrasio goes beyond headless browsers by patching the Chrome binary for fingerprint spoofing and routing through residential IPs, making it suitable for sites that detect and block conventional automation.

## Website
- Homepage: https://www.scrapetechnology.com
- Contact: https://www.scrapetechnology.com/contact
- Discord: https://discord.gg/GBSKsC8DvS