The schema skill guides an agent through structured data work. It asks about page type, current markup and target rich results, then produces complete JSON-LD for types such as Organization, Article, Product, SoftwareApplication, BreadcrumbList, LocalBusiness and Event.
It tells the agent to check Google’s current supported feature list before writing markup, and records that FAQ and HowTo rich results and the sitelinks search box have been retired. A testing checklist covers the Schema.org Validator, the Rich Results Test and Search Console.
What You Get
- A complete JSON-LD code block for the page type in question
- A check of whether the requested Google search appearance is still supported
- A multi-type @graph that combines Organization, WebSite and BreadcrumbList
- A testing checklist covering required fields, content match and validator results
- Implementation guidance for a static template, a server-rendered component or a CMS plugin
Common Use Cases
- Adding Product and Offer markup to pricing or product pages
- Fixing missing-property errors reported in Search Console
- Deciding whether to add FAQPage markup after Google retired its rich results
- Adding Organization and BreadcrumbList markup across a new site
What Makes It Unique
It separates valid Schema.org vocabulary from eligibility for a Google feature, and states the retirement dates for FAQ, HowTo and the sitelinks search box so the agent does not promise a rich result that no longer exists. It also says not to treat a missing report for a retired feature as a broken implementation.
Under The Hood
The skill is an instruction file with one reference document of JSON-LD examples. It has no scripts. The agent reads the product marketing context file if present, asks about page type and tech stack, and writes markup in chat. It names Google’s Rich Results Test and the Schema.org Validator for the user to run, and links to Google documentation, but it does not fetch pages, call APIs, run commands or write files itself.