Generated sitemap.ts and robots.ts

Two TypeScript files generate the artefacts crawlers read: app/sitemap.ts produces /sitemap.xml and app/robots.ts produces /robots.txt. They are routes like any other, with the same caching limits.

The sitemap

// app/sitemap.ts
import type { MetadataRoute } from "next";

const BASE = "https://ui.eneawork.it";
const PAGES = ["", "/pricing", "/blog"];

export default function sitemap(): MetadataRoute.Sitemap {
  return PAGES.map((path) => ({
    url: `${BASE}${path}`,
    lastModified: new Date(),
    changeFrequency: "weekly" as const,
    priority: path === "" ? 1 : 0.7,
    alternates: {
      languages: { it: `${BASE}/it${path}`, en: `${BASE}/en${path}` },
    },
  }));
}

The MetadataRoute.Sitemap type accepts url, lastModified, changeFrequency, priority and alternates.languages. The alternates field writes the xhtml:link entries for hreflang: without it the sitemap stays valid but says nothing about translations.

The robots file

// app/robots.ts
import type { MetadataRoute } from "next";

export default function robots(): MetadataRoute.Robots {
  return {
    rules: [{ userAgent: "*", allow: "/", disallow: ["/api/", "/admin/"] }],
    sitemap: "https://ui.eneawork.it/sitemap.xml",
    host: "https://ui.eneawork.it",
  };
}

The 50,000 cap

One sitemap file holds at most 50,000 URLs and 50 MB uncompressed. Past that threshold you use generateSitemaps, which returns an array of ids and leaves one file per id to you.

// app/sitemap.ts
export async function generateSitemaps() {
  return [{ id: 0 }, { id: 1 }, { id: 2 }];
}

export default async function sitemap({ id }: { id: number }) {
  const rows = await loadUrls();
  return rows.slice(id * 50000, id * 50000 + 50000);
}

Static or dynamic

The routes that end up in a sitemap are pages: blog blocks, pagination and breadcrumbs cover the archive without inventing URLs.

Index or single file is in sitemap index vs a single file; submission to Google in Search Console and indexing; internal links in an internal linking strategy. Comparison: Next.js vs Vite for a marketing site.

curl -s http://localhost:3000/sitemap.xml

Known limitation: a sitemap declares URLs, it does not check them. A page.tsx removed without dropping its entry produces a 404 that Search Console reports; the check belongs in CI, not in your eyes.