In a Warlock app

@warlock.js/sitemap remains framework-blind. In a Warlock application, @warlock.js/web discovers page policy, publishes sitemap generations, and serves the HTTP routes.

Install and configure

warlock add sitemap

The command adds a disabled section in src/config/web.ts. Enable it after app.publicUrl is configured:

src/config/web.ts

import type { WebConfigurations } from "@warlock.js/web";

const web: WebConfigurations = {
  sitemap: {
    enabled: true,
    storage: { directory: "sitemap" },
    coordination: "local",
  },
};

export default web;

The origin is app.publicUrl; no request-derived fallback is used. Missing origin makes generation fail safely and /sitemap.xml returns 503 until a generation succeeds.

storage selects durable generation and manifest storage. Its directory must be a dedicated relative prefix. Omit disk for the app’s default storage. outputDir is the compatible local writer option; it conflicts with storage rather than silently winning.

What Web adds

Web includes static pages, applies config.sitemap: false, expands locales, and lets dynamic pages declare concrete URLs with config.sitemap.entries. It keeps each published generation immutable, serves the latest valid manifest, and never creates a document in response to a crawler request.

Use coordination: "shared" only with storage that supports atomic create-if-absent and consistent reads/listing. No Redis service is required. See the Sitemap and robots.txt guide for refreshes, model invalidation, conditional HTTP caching, shards, and robots.txt.

API-only applications can use new Sitemap({ baseUrl }) directly and choose their own delivery route. See Build a sitemap.