Ivan Hawkes

No structured data, and thin social meta tags.

Current State

Measured over public/ of the current build. be-best/layouts/_partials/meta/meta.html writes a canonical link, a description, og:title, og:description, og:url, og:type, twitter:card, twitter:title and twitter:description, plus og:image, twitter:image and twitter:site when the params behind them are set. The gaps:

Structured data is absent: no page carries an application/ld+json block, so the structured data audit scores 0/100 and no page is eligible for rich results.
og:image and twitter:image are written only when .Site.Params.socialImage is set, and config/_default/params.yaml sets no socialImage, so a shared link carries no image. twitter:site emits @ivanhawkes, read from social.twitter, added on 2026-10-29.
og:type is hard-coded to website on every page, so posts and recipes carry no article properties (article:published_time, article:author, article:section).
There is no robots, author, publisher, creator or keywords meta tag, which is what holds the meta tags score at 74/100.
sitemap.xml and one index.xml feed are already generated, and the head links the feed: the crawl route exists, the descriptions of the pages do not.

Work

Params - define socialImage in config/_default/params.yaml, and add the share image to static/ through Git LFS: 1200x630 suits twitter:card summary_large_image.
Types - select og:type by content type: article for post pages, website for the home page and section pages, and the matching properties (published, modified, author, section) on articles.
JSON-LD - a head partial emitting, per page: WebSite on the home page, BlogPosting on a post (author Person, publisher, image, datePublished and dateModified), Recipe on type: recipe reading the ingredients.yaml sidecar and the prep-time / cook-time ISO durations, BreadcrumbList on every page below the home page, and AboutPage for the about/ pages.
Meta tags - add robots, author, publisher, creator and keywords (from the page tags), and state the indexing decision for the working sections: kanban and specification are working documents, so say whether they are indexed or noindex.

Acceptance Criteria

Every page carries a JSON-LD block of the type above, and the blocks validate against Google’s Rich Results Test with no errors; the warnings are listed on this card
og:image and twitter:image carry a real image on the home page and on a post, and a shared link shows it
A post page reports og:type article with published and modified times, an author and a publisher
The robots, author, publisher, creator and keywords tags are present, and the indexing decision for kanban and specification is visible in the built pages
The re-run review reports the structured data and meta tags scores, recorded on this card