Skip to content
Nibble

Building a theme

SEO, sitemaps and feeds

What search engines, social networks and feed readers receive from a Nibble site, and the settings that shape it.

A startup’s site earns most of its visitors from search and from links people share. This guide covers everything Tidewater’s site tells search engines, social networks and feed readers — most of which needs no work at all.

After reading this guide, you will know:

  • Where a page’s title, description and share image come from.
  • How the SEO global sets defaults for the whole site.
  • How sitemaps and robots.txt are generated, and when a site asks not to be indexed.
  • How to publish Atom feeds.

1. Page titles, descriptions and share images

Every blueprint that imports nibble::seo gives editors an SEO tab: a title, a description, a share image, a canonical URL, a “hide from search engines” switch, and the kind of structured data the page is.

Left empty, each falls back to something sensible:

Tag Comes from
title the SEO title, else the entry’s title — then put through the site’s title template
description the SEO description, else the entry’s excerpt, else the site’s default description
share image the SEO image, else the featured image, else the site’s default share image
canonical URL the SEO canonical URL, else the page’s own URL

So a post published without anyone touching the SEO tab still has a proper title, description and share card.

The page gets its title, description, canonical link and robots tag from SeoHead, and its Open Graph and Twitter tags, structured data and verification tags from the HTML Nibble sends. See Theme components.

2. Site-wide defaults

Globals → SEO holds the defaults:

Field Tidewater’s
Title template {title} · Tidewater — tokens are {title} and {site_name}
Default description “Invoicing software that gets small agencies paid.”
Default share image a card with the logo
Organization name and logo used in structured data
Verification tags the tags Google Search Console and others ask you to add

3. Sitemaps

/sitemap.xml is an index pointing at one sitemap per collection and taxonomy with a route — /sitemap-posts.xml, /sitemap-help.xml, /sitemap-taxonomy-topics.xml. They list published content only, stay current as content is published, and include pages written as files.

Tune a collection’s entry, or leave it out:

# site/schema/collections/posts.yml
sitemap:
  priority: 0.8
  changefreq: weekly

sitemap:
  enabled: false

An entry whose SEO tab says “hide from search engines” is left out of the sitemap and marked noindex.

4. robots.txt, and sites that should not be indexed

/robots.txt is generated. On the production site it allows everything and names the sitemap:

User-agent: *
Allow: /

Sitemap: https://tidewater.example/sitemap.xml

Anywhere else — development, or production with NIBBLE_BLOCK_INDEXING set — it disallows everything, every page is marked noindex, nofollow, and analytics stay off.

Important

Set NIBBLE_BLOCK_INDEXING=1 on any public server that is not the real site: a demo, a preview, a copy for testing. Otherwise it competes with your real site in search results.

5. Feeds

A collection with feed publishes an Atom feed of its newest published entries:

# site/schema/collections/posts.yml
feed:
  title: The Tidewater blog
  limit: 20
URL Holds
/feed-posts.xml the blog
/feed.xml every collection with a feed, newest first

Each item carries the entry’s title, URL, date and — when the entry has one — its excerpt or description as the summary. feed: true uses the collection’s title and the newest 50 entries.

Note

Feeds read entries from the database, so a collection written as files has no feed.

6. Redirects and missing pages

Moving pages without losing their search ranking is covered from the editor’s side in Redirects and SEO: changing a live entry’s slug records a 301 automatically, chains are collapsed to a single hop, and every 404 is counted so it can be redirected.

Next
The Content API