Schema markup

Skill tmj-90/gaffer/runner/skills/schema-markup

Self-hosted AI coding factory — sandboxed agents deliver tickets to merged code, gated by a human in a dashboard. Local-first, cost-transparent, human-in-the-loop.

Install

npx -y skills add tmj-90/gaffer --skill schema-markup

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Use when implementing, auditing, or validating structured data (schema markup) on a website. Triggers on "structured data", "schema.org", "JSON-LD", "rich results", "rich snippets", "FAQ schema", "Product schema", "schema errors in Search Console", or "why no rich results". NOT for general SEO audits — use `seo-audit`. For AI-search citation optimisation, use `aeo`.

SKILL.md

4.3 KB, 936 tokens by cl100k_base, as published. Nobody here has run it

Implement structured data that earns rich results

JSON-LD is the recommended format (Google's preference). Schema.org types are the vocabulary. Rich results in Google and citations in AI search engines are the outcomes.

High-value schema types (implement these first)

Type	Rich result earned	Required fields
`Organization`	Knowledge panel, sitelinks search	`name`, `url`, `logo`
`WebSite`	Sitelinks search box	`url`, `potentialAction` (SearchAction)
`FAQPage`	FAQ dropdowns in SERP	`mainEntity` → `Question` → `acceptedAnswer`
`Article`	Top stories, rich snippet	`headline`, `author`, `datePublished`, `image`
`Product`	Product snippet with price/rating	`name`, `offers`, `aggregateRating`
`HowTo`	How-to steps in SERP	`name`, `step` array
`BreadcrumbList`	Breadcrumb in SERP URL	`itemListElement` array
`LocalBusiness`	Local pack, knowledge panel	`name`, `address`, `telephone`

JSON-LD placement

Prefer <head> as a <script type="application/ld+json"> block. Google also supports JSON-LD placed in the <body> (including markup injected by JavaScript), so in-body placement is valid when <head> injection isn't practical — <head> is simply the cleaner default. One <script> block per schema type per page (or a @graph array for multiple types).

<script type="application/ld+json">
{
  "@context": "https://schema.org",
  "@type": "FAQPage",
  "mainEntity": [
    {
      "@type": "Question",
      "name": "What is your refund policy?",
      "acceptedAnswer": {
        "@type": "Answer",
        "text": "We offer a 30-day full refund, no questions asked."
      }
    }
  ]
}
</script>

Common mistakes that block rich results

Content mismatch — schema claims a price or rating that doesn't appear on the page. Google requires visible content to match.
Missing required fields — check Google's required/recommended fields per type at developers.google.com/search/docs/appearance.
Incorrect datePublished format — must be ISO 8601 (2026-06-17T09:00:00+00:00).
Markup on 404/noindex pages — Google won't process structured data on non-indexable pages.
Multiple conflicting types — use @graph array to combine without conflict.

Steps

Audit existing markup. Paste page source into Google Rich Results Test; review Search Console → Enhancements for errors. Identify what's present, what's broken, what's missing.
Prioritise by page type. Homepage → Organization + WebSite. Blog posts → Article. FAQs → FAQPage. Product pages → Product. All inner pages → BreadcrumbList.
Implement in JSON-LD. Populate all required fields; add recommended fields for richer results. Confirm all schema claims match visible page content.
Validate. Run through Google Rich Results Test (schema.googleapis.com/v1/richResults); fix any errors (not just warnings).
Monitor. After deployment, check Search Console → Enhancements in 48–72 hours for indexation of new schema.
Record evidence. Screenshot Rich Results Test passing; note which enhancements are expected.

Review checklist

JSON-LD in <head> — not inline or in body.
All required fields present — per Google's type documentation.
Content matches page — every schema claim visible on the rendered page.
Rich Results Test passes — zero errors (warnings acceptable).
BreadcrumbList on all inner pages — not just the homepage.
Search Console errors cleared — no existing structured data errors before adding new.

Rules

Schema that describes content not visible on the page is spam — Google may penalise it.
Fix Search Console errors on existing schema before adding new types.
Never use Microdata or RDFa — JSON-LD only (maintainable and Google's preference).

What ships with it

Read from the repository

Just SKILL.md. No reference files, no scripts.