Get started

ScrapeFlow: the web data API

Fast web scraping with browsers, proxies, retries, batches, and monitors built in — on a serverless backend that costs less to run, so it costs you less to use.

Get data from the internet, fast. ScrapeFlow handles the scraping infrastructure so you can turn web pages into Markdown, structured JSON, screenshots, and more through one API. Start with one URL, scrape thousands of pages in a batch, or monitor sites for changes.

✓

Serverless by design. Every request runs on AWS Lambda with managed proxies and JavaScript rendering on by default. You pay per request — never for idle capacity. See Pricing & credits for the real unit economics.

⌘

Quickstart

Get an API key and make your first request.

🤖

Agent quickstart

Paste one prompt into your coding agent to set up ScrapeFlow.

</>

API reference

Every endpoint's parameters and response fields.

Make a request

Every request is authenticated with a bearer token. Set your key as SCRAPEFLOW_API_KEY and the base URL as SCRAPEFLOW_API_URL, then scrape your first page:

curl -X POST $SCRAPEFLOW_API_URL/v1/web/scrape \
  -H "Authorization: Bearer $SCRAPEFLOW_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{ "url": "https://linear.app", "formats": { "markdown": true } }'

Choose an API

ScrapeFlow is one platform with several surfaces. Pick the one that fits your workflow — they all share a single API key and credit balance.

• Scrape — one URL → Markdown, HTML, JSON, or a screenshot.
• Search — query the web and get ranked results.
• Answers — sourced, structured research answers.
• Crawl — an entire site, async, on SQS workers.
• Map — discover every URL on a domain.
• Brand — a typed company + design profile from a domain.
• Batches — thousands of URLs in one fan-out job.
• Monitors — scheduled change detection with webhooks.

Before you ship

Keep your API key server-side, handle 429 responses with exponential backoff, and set a monthly budget cap in the dashboard so a runaway job can never surprise your bill.