@nukipa/site-sdk

Typed client for the Nukipa public API (/public/v1/*). Used by tenant sites under sites/ to fetch content and ship signals back.

import { createNukipaClient } from '@nukipa/site-sdk';
import { headers } from 'next/headers';

const client = createNukipaClient({
  gatewayUrl: process.env.NUKIPA_GATEWAY_URL!,
  // Resolved per-call. Server components / RSC can read from the request.
  getHost: async () => process.env.NUKIPA_TENANT_HOST
    || (await headers()).get('x-forwarded-host')
    || (await headers()).get('host')
    || ''
});

const tenant = await client.getTenant();
const posts  = await client.listPosts({ limit: 10 });
await client.recordVisit({ path: '/blog' });

Source of truth

The TypeScript types in src/types.gen.ts are generated from the OpenAPI spec at apps/gateway/openapi.public.yaml. Update the YAML when you add or change a public endpoint, then regenerate:

npm run codegen -w @nukipa/site-sdk

The runtime wrapper in src/client.ts is hand-maintained — codegen only touches types.gen.ts. When adding a new operation, you'll edit both: the spec for the type and the client method that calls it.

Why hand-written runtime?

openapi-fetch would buy us auto-typed methods, but the public surface is small (~9 ops) and the per-method ergonomics we want (host resolution, swallowed analytics errors, next.revalidate defaults) are nicer with a small hand-rolled wrapper. If the surface grows past ~20 endpoints, revisit.