What is Scrapio?
Scrapio lets AI agents, automation pipelines, and developers extract structured content from the web through a single HTTP API. You don’t manage browsers, proxies, or rendering infrastructure — we do.Surfaces
Fetch
Fetch and render a single URL. Returns HTML, Markdown, JSON, or a screenshot.
Crawl
Crawl a site starting from one or more seed URLs. Controls depth, domain scope, and page budget.
Extract
Pull structured data from any page using CSS selectors, a JSON schema, or a natural-language instruction.
Interact
Drive a real browser: click, type, scroll, wait for elements, take screenshots.
Search
Run a Google search and optionally fetch each result page in the same call.
YouTube
Search YouTube, retrieve video metadata, and download subtitles/transcripts.
MCP Server
Use the Scrapio directly from Claude, Cursor, Windsurf, and any MCP-compatible agent runtime — no HTTP calls required.MCP Server
One-line install for Claude Desktop, Cursor, and Windsurf. Nine tools covering fetch, search, crawl, and browser automation.
Async jobs
Every surface supports async execution. Submit a job, get back ajob_id, and poll /v1/jobs/{id} for the result. Long crawls and multi-step interactions use this path automatically.
Before you start
You need an API key. Get one at scrapio.dev — the free tier includes 1,000 credits per day. All requests are authenticated withAuthorization: Bearer <api-key>.
Quickstart
Make your first API call in under 60 seconds.