Introduction

Build with Orsa.dev

Orsa is the Internet’s Context API — web data extraction and brand intelligence in a single platform.

What is Orsa?

Orsa provides a unified API for:

  • Web Scraping — Extract markdown, HTML, plain text, images, and sitemaps from any URL
  • Brand Intelligence — Get logos, colors, fonts, social links, and industry data from any domain
  • AI Extraction — Ask natural language questions about any website or pull structured product lists
  • Screenshots — Capture a website’s homepage as an inline base64 PNG
  • Industry Classification — NAICS-style industry tags for any indexed brand

Quick Example

import Orsa from '@orsa.dev/sdk';
 
const client = new Orsa({ apiKey: process.env.ORSA_API_KEY! });
 
// Get brand data for any company. Returns the Context.dev shape — colors and
// logos are arrays of objects (with `*_legacy` dict mirrors for older code).
const brand = await client.brand.retrieve({ domain: 'stripe.com' });
console.log(brand.title);              // "Stripe"
console.log(brand.colors[0]?.hex);     // "#635BFF"
console.log(brand.logos[0]?.url);      // logo URL
console.log(brand.socials);            // [{ type: "github", url: "https://github.com/stripe" }]
 
// Scrape a page. Default mode is markdown; pass 'html' or 'text' for those.
const page = await client.web.scrape({ url: 'https://stripe.com/pricing' });
console.log(page.markdown);
 
// AI-powered data extraction.
const ans = await client.ai.query({
  domain: 'stripe.com',
  dataToExtract: 'Return the pricing plans as a JSON array.',
});
console.log(ans.result);

Every method returns the unwrapped payload — you don’t need to reach into .data yourself.

Key Features

FeatureDescriptionCredits
Brand RetrieveFull brand data from a domain10
Brand Retrieve (Simplified)Just domain/title/logo/primaryColor/industries10
Brand Retrieve by NameFuzzy search returning top match + candidates2
Scrape (markdown/html/text)Single-page extract1
Scrape ImagesImages with role classification1
Scrape SitemapURLs + path-grouped buckets1
ScreenshotInline base64 PNG5
FontsFamily + source extraction5
NAICS ClassificationIndustry tags5
Transaction IdentifierMatch bank descriptors to brands10
StyleguideW3C-DTCG tokens + DESIGN.md15
AI QueryNatural language extraction20
AI ProductsTool-use structured product list15

Getting Started

  1. Create an account and get your API key
  2. Follow the Quickstart Guide
  3. Explore the TypeScript SDK and API Reference

Open Source

Orsa’s core engine is fully open source. Self-host, extend, or contribute on GitHub.