Skip to content
ThinkTools

Scrapia

Point-and-click web scraper with self-healing selectors, a CLI, an API and a scheduler

  • Point-and-click recorder
  • Self-healing selectors
  • 17 export formats and destinations
  • Desktop, CLI, REST API and scheduler
  • MCP server for AI agents
Buy now · $39.99See pricing

One-time payment · Delivered as soon as payment confirms · No added fees

  • Windows
  • macOS
  • Linux
  • Docker
  • Web

Screenshots

Scrapia in use

Run monitor with live counters1 / 5

Screenshot

Overview

About Scrapia

Scrapia is a visual web scraper. You open a site in its recording browser and click the items you want; Scrapia detects the repeating records, proposes fields, generates ranked selectors and shows the data live. The result is a recipe, a JSON file you can run on your desktop, from the command line, on a server through a REST API, on a schedule, or from an AI agent over MCP.

Every field keeps three to five ranked selector candidates. When a site changes, the engine falls through to the next candidate, logs the drift and can ask a language model to repair the selector, verifying the fix before it is used. You get an alert instead of a silently broken scrape.

Pages are fetched over HTTP first and opened in a stealth Chromium browser only when they need JavaScript or present a challenge. Recipes handle next links, numbered pages, load-more buttons, infinite scroll, detail pages, multi-level categories, keyword lists and logins, with per-host rate limits, proxy pools and CAPTCHA solver plugins.

Data comes out typed and clean, with transforms, validation, de-duplication and upserts by key, and exports to CSV, Excel, JSON, Parquet, SQLite, PostgreSQL, MySQL, MongoDB, Google Sheets, Airtable, Notion, S3 and webhooks. Monitoring compares each run with the previous one and alerts you on Slack, Discord, Telegram, ntfy, email or a webhook. An assistant panel in the recorder can build a recipe from a plain-language description, using a cloud model or a local one through Ollama or LM Studio.

Features

What it does

Build a scraping recipe by clicking in a real browser, then run it on your desktop, from the command line, on a server, on a schedule or from an AI agent.

  • Visual recorder

    Click elements in a real browser and Scrapia finds the records, suggests fields and previews the data live.

  • Self-healing selectors

    Ranked fallbacks for every field, drift logging and optional verified repairs keep scrapes running when sites change.

  • HTTP first, browser when needed

    Fast HTTP fetching that upgrades to stealth Chromium for JavaScript pages, with proxy pools and rate limits.

  • Every navigation pattern

    Pagination, load more, infinite scroll, detail pages, categories, keyword lists and login flows.

  • Clean, typed data

    Regex, number, currency and date transforms, validation rules, and de-duplication or upserts by key.

  • Every export

    CSV, Excel, JSON, Parquet, SQL databases, MongoDB, Google Sheets, Airtable, Notion, S3 and signed webhooks.

  • Change monitoring

    Compare each run with the last one and get alerts on Slack, Discord, Telegram, ntfy or email.

  • Run anywhere

    Desktop app, command line, REST API with live events, a cron scheduler and a Docker image.

  • AI agents over MCP

    An MCP server exposes recipes, runs and datasets as tools for AI agents, and an assistant can draft recipes for you.

Pricing

One price, everything included

Pay once in USD with crypto. Your download links and license key arrive by email as soon as the payment confirms.

Full source code

Visual scraper, recorder, engine, CLI, API, scheduler and MCP server.

$39.99

One-time payment · USD · no fees added

Buy now

What you get

  • Complete source code for every component
  • Desktop recorder, headless engine, CLI and REST API
  • Scheduler and MCP server
  • All future updates at no extra cost
  • Email support

Every purchase includes

  • Instant delivery

    Download links and your license key by email as soon as payment confirms.

  • Free future updates

    Every future version at no extra cost, on your order page the day it ships.

  • Security reviewed

    Tested by our engineers and by AI-driven analysis before every release.

  • Pay with crypto

    The payment page shows the exact amount, the address and a QR code.

  • No added fees

    You pay exactly the listed price.

  • Support from real people

    Questions go to humans@thinktools.dev.

FAQ

Questions and answers

Can't find what you're looking for? See the store FAQ or write to humans@thinktools.dev.

What do I receive?
Complete source code for every component: the desktop recorder, the headless engine, the command line tool, the REST API server, the scheduler and the MCP server, plus install and server deployment guides, all updates and email support.
Which systems does it run on?
The desktop app runs on Windows, macOS and Linux. The engine, command line tool and API need Node.js 22.13 or later and run on any server, or in the included Docker image.
Do I need an AI subscription?
No. Recording, running, scheduling and exporting work without any model. The optional assistant, AI field extraction and selector repair work with a cloud API key or with a local model through Ollama or LM Studio, so your data can stay on your machine.
Can it handle logins and CAPTCHAs?
Recipes support login flows with persistent browser profiles. CAPTCHAs can be passed to 2Captcha, CapSolver or Anti-Captcha plugins, or solved by hand in a visible browser window.
Is it my responsibility to scrape responsibly?
Yes. Follow each site's terms and the laws that apply to you. Scrapia gives you per-host rate limits, randomized delays and a robots.txt switch to help you stay polite.
How do updates, support and refunds work?
Updates are delivered through your order page, and email support comes from real people at humans@thinktools.dev. For refund terms, see Refunds.