Orthogonal - The Unified API for AI Agents

Agents building onboarding flows, CRM enrichment, or competitor analysis need structured data about companies. Logos, colors, descriptions, social handles, industry codes, product catalogs. The usual approach is scraping a website, parsing HTML, guessing at image URLs, and hoping nothing breaks when the site redesigns. It's fragile, slow, and every company's site is different.

Today we're excited to announce our partnership with Context.dev, a platform that turns any domain or URL into structured, AI-ready data with a single API call. Send a domain, get back logos in every format, brand colors, company metadata, social handles, industry classification, and more. Send a URL, get clean markdown, a full design system, product data, or a screenshot. Every response is typed JSON. Context.dev powers brand data for companies like Mintlify, daily.dev, and Propane.

What is Context.dev?

Context.dev provides five categories of APIs that extract structured information from domains and URLs:

Brand Intelligence: Retrieve logos (light, dark, icon variants in SVG + PNG), brand colors, company profiles (description, slogan, email, phone), social handles for 25+ platforms, company addresses, and stock tickers. Look up brands by domain, company name, email address, stock ticker, or ISIN.
Web Scraping: Scrape any URL to clean markdown or raw HTML, crawl entire sitemaps, capture screenshots at custom viewports, search the web with optional markdown extraction, and extract images with metadata.
AI Extraction: Extract product data (name, price, features, images, SKU) from any product page, and extract custom data points from any website by describing what you need.
Design Systems: Pull full design systems (colors, typography, spacing, shadows, component styles) and scrape font families with usage stats from any brand's website.
Classification: Classify any brand into NAICS codes with confidence scores. Enrich transaction descriptors (like "AMZN MKTP US") into full merchant profiles with logos and industry labels.

Bot detection bypass and proxy escalation are handled automatically. No browser automation required on your end.

Using Context.dev with Orthogonal

Retrieve Brand Data

Get a complete brand profile from any domain: logos, colors, description, social handles, address, and industry classification.

import Orthogonal from "@orth/sdk";

const orthogonal = new Orthogonal({
  apiKey: process.env.ORTHOGONAL_API_KEY,
});

const result = await orthogonal.run({
  api: "context-dev",
  path: "/brand/retrieve",
  query: {
    domain: "stripe.com"
  }
});

console.log(result);
// Returns: logos (SVG + PNG, light/dark/icon variants),
// brand colors, title, description, slogan, social handles,
// address, stock ticker, NAICS industry classification, and more

You can also look up brands by company name, email address, or stock ticker:

const result = await orthogonal.run({
  api: "context-dev",
  path: "/brand/retrieve-by-name",
  query: {
    name: "Stripe"
  }
});

Scrape to Markdown

Turn any URL into clean, LLM-ready markdown with navigation and ads stripped out.

const result = await orthogonal.run({
  api: "context-dev",
  path: "/web/scrape/markdown",
  query: {
    url: "https://stripe.com/pricing"
  }
});

console.log(result.markdown);
// Returns: clean markdown content ready for RAG pipelines,
// summarization, or agent consumption

Extract Products

Given a product page URL, extract structured product data: name, description, price, billing frequency, features, images, category, and SKU.

const result = await orthogonal.run({
  api: "context-dev",
  path: "/brand/ai/product",
  method: "POST",
  body: {
    url: "https://allbirds.com/products/mens-wool-runners"
  }
});

console.log(result);
// Returns: product name, description, price, currency,
// billing frequency, pricing model, features, images,
// category, target audience, tags, and SKU

Extract Structured Data with a Schema

Define the shape of data you want as a JSON Schema, and Context.dev crawls the site and fills it in.

const result = await orthogonal.run({
  api: "context-dev",
  path: "/web/extract",
  method: "POST",
  body: {
    url: "https://linear.app",
    schema: {
      type: "object",
      properties: {
        has_free_trial: { type: "boolean" },
        pricing_tiers: {
          type: "array",
          items: {
            type: "object",
            properties: {
              name: { type: "string" },
              price: { type: "number" }
            }
          }
        },
        founded_year: { type: "number" }
      }
    },
    instructions: "Find pricing and company info"
  }
});

console.log(result.data);
// Returns typed JSON matching your schema:
// { has_free_trial: true, pricing_tiers: [...], founded_year: 2019 }

Industry Classification

Classify any brand into NAICS codes from just a domain or company name.

const result = await orthogonal.run({
  api: "context-dev",
  path: "/web/naics",
  query: {
    input: "shopify.com"
  }
});

console.log(result);
// Returns: NAICS code, industry name, and confidence score

Using x402 Protocol

Context.dev on Orthogonal also supports x402, an open protocol that enables native payments in HTTP. Your agents can pay for API calls directly using USDC stablecoins, no API keys required.

import { wrapFetchWithPayment } from "x402-fetch";
import { privateKeyToAccount } from "viem/accounts";

const privateKey = process.env.PRIVATE_KEY;
if (!privateKey) {
  throw new Error("PRIVATE_KEY environment variable is required");
}
const account = privateKeyToAccount(
  privateKey.startsWith("0x") ? privateKey : `0x${privateKey}`,
);
const fetchWithPayment = wrapFetchWithPayment(fetch, account);

const response = await fetchWithPayment(
  "https://x402.orth.sh/context-dev/brand/retrieve?domain=stripe.com",
  { method: "GET" }
);

const data = await response.json();
console.log(data);

Using MPP (Machine Payments Protocol)

Context.dev is also available via MPP, the open standard for machine-to-machine payments co-authored by Tempo and Stripe. MPP supports stablecoins, cards, and Bitcoin on the same endpoint.

import { privateKeyToAccount } from "viem/accounts";
import { Mppx, tempo } from "mppx/client";

Mppx.create({
  methods: [tempo({ account: privateKeyToAccount("0x...") })],
});

const response = await fetch(
  "https://mpp.orthogonal.com/context-dev/brand/retrieve?domain=stripe.com"
);

const data = await response.json();
console.log(data);

The mppx client handles 402 challenges automatically: it intercepts the payment request, signs a credential with your wallet, and retries with proof of payment.

Capabilities

Context.dev's APIs cover a wide surface:

Brand lookups by domain, name, email, stock ticker, or ISIN with logos in SVG + PNG (light, dark, and icon variants)
25+ social platform handles per brand (X, LinkedIn, GitHub, YouTube, and more)
NAICS industry classification with confidence scores
Transaction enrichment that resolves raw card descriptors to real merchants
Full design system extraction: colors, typography, spacing, shadows, and component CSS
Product extraction with structured fields for name, price, features, images, and SKU
Schema-driven extraction that lets you define exactly what data you want from any website
Web scraping to markdown, HTML, or screenshot with automatic bot detection bypass
Site crawling across entire sitemaps with markdown extraction for every page

Use Cases

CRM and Onboarding Enrichment

When a user signs up with a work email, resolve their domain to get company logos, brand colors, industry, and social links. Pre-fill onboarding forms and personalize their experience from day one.

Agent-Driven Research

Agents building competitive analyses can scrape competitor sites to markdown, extract pricing data, pull product catalogs, and classify companies by industry. All structured, all typed.

Transaction Categorization

Fintech platforms can turn raw card descriptors like "AMZN MKTP US" into full merchant profiles with logos, colors, and industry labels. Build spend analytics and transaction tagging without maintaining a merchant database.

White-Label Theming

Extract a customer's full design system (colors, fonts, spacing, component styles) from their website and auto-generate white-labeled UIs that match their brand.

Logo Walls and Partner Pages

Use Logo Link to embed always-fresh brand logos from a CDN without API calls. Build dynamic logo walls, partner pages, and integration directories that stay current automatically.

Why We Partnered with Context.dev

Most approaches to extracting company data from the web involve brittle scrapers, regex, and constant maintenance. Context.dev replaces all of that with typed API responses that handle bot detection, proxy escalation, and data normalization for you. Whether your agent needs a company logo, a pricing page in markdown, or a full product catalog, it's one API call that returns structured JSON. That reliability and breadth is exactly what agents need to operate autonomously.

Try It Today

Sign up for Orthogonal and get $10 free credits to try Context.dev and dozens of other APIs. One key, hundreds of APIs, pay per request.

Get Started | View on Orthogonal | Context.dev Website

Context.dev: Turn Any Domain into Structured, AI-Ready Data