Scrape SDK

TypeScript Scrape SDK

Scrape, search, crawl, extract, and optionally run web agents with one TypeScript client across Firecrawl, TinyFish, Jina, Tavily, Spider, Browserbase Fetch, and local Cheerio.

View Markdown

Scrape SDK is a provider-neutral web client for TypeScript agents. It talks to real vendor APIs, fails over on retryable errors, and returns one typed result shape.

scraper.ts
import { scrape } from "scrape-sdk";

const page = await scrape("https://stripe.com/docs");
console.log(page.markdown);
console.log(`via ${page.provider} in ${page.latencyMs}ms`);

Use search() when you do not have a URL, crawl() for a site, and extract() for JSON Schema. Providers that cannot perform an operation are skipped. A client (fromEnv() / createScrapeClient) is optional — scrape(url) is enough.

On this page