Set canonicals and sitemap with Astro integrations

Why this matters

One config value makes URL and sitemap generation trustworthy.

Before you start

• An Astro project with a production domain
• Ability to run a production build locally or in CI

How to apply it

1. Set the production site

   In astro.config.mjs set site: 'your production domain' using the production origin with no trailing slash. This value feeds canonicals, RSS links and sitemap entries.

   Check: grep -n 'site:' astro.config.mjs shows the production origin and astro build finishes without a site warning.

2. Install the sitemap integration

   Run npx astro add sitemap, which installs @astrojs/sitemap and adds it to integrations. Build and confirm sitemap-index.xml and its children land in dist/.

   Check: dist/sitemap-index.xml exists after astro build and contains only absolute https URLs.

3. Render a canonical in the layout

   In src/layouts/Base.astro compute const canonical = new URL(Astro.url.pathname, Astro.site) and add link rel=canonical href={canonical} in the head so each route declares itself.

   Check: curl -s your production domain/ | grep -c 'rel="canonical"' returns 1 with an absolute href.

4. Filter drafts and previews

   Use the filter option of the sitemap integration to drop draft, preview and thin archive routes, while keeping paginated pages that are meant to be indexable.

   Check: The built sitemap contains no preview or draft URLs and every loc returns 200.

5. Deploy and submit

   Deploy the build, request sitemap-index.xml in production and add it in Search Console. Confirm robots.txt links to the same index rather than an old sitemap file.

   Check: curl -s your production domain/sitemap-index.xml returns 200 and Search Console reports the sitemap as Success.

<goal>
link alt index report layout heading signal heading crawl serp redirect heading anchor query.
</goal>

<context>
index alt snippet crawl heading crawl verify audit signal audit template crawl locale crawl.
</context>

<discover_first>
heading measure redirect metadata page redirect schema target query target locale locale schema locale.
</discover_first>

<constraints>
heading route intent crawl caching metadata locale anchor markup snippet serp metadata link audit.
</constraints>

<plan>
query content canonical audit target audit verify sitemap link priority snippet intent verify section.
</plan>

<verify>
query priority markup crawl intent sitemap locale heading report link layout cluster metadata route.
</verify>

<report>
cluster priority target snippet priority crawler serp template breadcrumb cluster audit image cluster alt.
</report>

Source: seo-mcp playbook · curated entry · report an error