Choose your integration path

Four ways to call Quravin — plain webpage, React SPA, Node.js server, or raw API with no SDK. Pick the one that matches your stack.

Every path below ends up calling the same API. The difference is where your code runs and what’s already in your stack — pick based on that, not on which sounds most advanced.

Which one is you?

Your situation	Guide
A plain HTML page (server-rendered, static, or a `<script>` tag dropped into any site) — no build step, no framework	Plain webpage direct integration
A React single-page app — you want the SDK wired into component state, not a raw script tag	React SPA integration
A Node.js backend — minting session tokens for a browser frontend, or calling the API directly from server-only code (a cron job, a webhook handler, a CLI)	Node.js server integration
Any other language, or you’d rather not add a dependency — you just want the exact HTTP requests	Direct API integration

Before any of these

You need credentials first — see Get your API key. Every path above needs either a client_id/client_secret pair (to mint session JWTs) or a static x-api-key; Choose your auth mode explains which one fits where your code runs.

The short version, if you’re in a hurry

Browser code (webpage or React) → session JWT, minted by your own backend. Never put a static x-api-key in code that ships to a browser.
Server-only code (Node.js backend, cron job, internal service) → either mode works; a static x-api-key is simpler if you don’t need per-user attribution.
No SDK at all → the Direct API integration guide works from any language, with both auth modes shown as plain curl examples.