HOME SKILLS BLOG GITHUB
// SKILL

SCHEMA MARKUP
DETECT. VALIDATE. GENERATE.

Schema markup is a standardized way to describe page entities and properties so search systems can interpret them. Claude SEO detects JSON-LD, Microdata, and RDFa, checks markup against applicable vocabulary and search features, flags retired types, and generates review-ready JSON-LD drafts.

By Daniel Agrici · Command surface verified: · Source: Google structured data guidance

$
/seo schema https://your-site.com

REQUIRES CLAUDE SEO INSTALLED IN CLAUDE CODE

A webpage maps into structured-data nodes, validation, and search-feature eligibility checks
// RUN PREVIEW

What will happen if you start this skill?

The command checks one page, separates generic vocabulary from Google feature eligibility, and writes review-ready drafts.

OUTCOME

Detected markup findings, validation guidance, and draft JSON-LD for supported opportunities.

REQUIRED INPUT

One reachable page supplied to /seo schema <url>, with visible facts available for verification.

READINESS

Server HTML is preferred. Blocked or client-rendered content can make retrieval incomplete and may require a supplied snapshot.

LOCAL OUTPUT

SCHEMA-REPORT.md and generated-schema.json, with verified values or clearly marked placeholders.

SIDE EFFECTS

Fetches the page and writes local files. It does not add markup to the live site or guarantee a rich result.

COST + APPROVAL

No separate paid provider is required. Verify visible-content parity, validate the final feature, and obtain approval before publication.

// HOW IT WORKS

How does schema analysis work?

When you run /seo schema, Claude SEO performs a three-phase analysis: detection, validation, and generation. It scans the page source for all three Schema.org formats, validates each block against Google's current requirements, and then generates any missing schema types that would benefit your page.

Per Google's December 2025 JavaScript SEO guidance, structured data injected via JavaScript may face delayed processing. For time-sensitive markup, especially Product and Offer schemas, the tool recommends including JSON-LD in the initial server-rendered HTML.

/seo schema <url> 1. DETECT JSON-LD, Microdata, RDFa 2. VALIDATE Google requirements 3. GENERATE missing schema types VALIDATION CHECKS @context present valid @type data types OK absolute URLs not deprecated SCHEMA-REPORT.md + generated-schema.json detection results + review-ready draft

What gets checked

The schema analysis covers detection of all three formats, validation against Google's current supported types, and generation of missing schema opportunities. It always recommends JSON-LD as the primary format, which is Google's stated preference.

3
FORMAT DETECTION
Scans for JSON-LD (script type application/ld+json), Microdata (itemscope, itemprop), and RDFa (typeof, property). Always recommends JSON-LD as Google's preferred format.
6
VALIDATION CHECKS
Checks for missing @context, invalid @type, wrong data types, placeholder text, relative URLs (should be absolute), and invalid date formats. Flags every issue with a fix.
20+
VOCABULARY AND FEATURES
Schema.org defines a broad vocabulary. Google's Search Gallery documents a narrower set of Search features with separate eligibility and property requirements.
7+
DEPRECATED DETECTION
Flags HowTo (Sept 2023), SpecialAnnouncement (July 2025), CourseInfo, EstimatedSalary, LearningVideo (June 2025), ClaimReview (June 2025), VehicleListing (June 2025), and more.
INFO
FAQPAGE STATUS
Google retired FAQ rich results for all sites on May 7, 2026. Existing FAQPage can remain for non-Google or internal semantics, but it has no Google rich-result benefit.
3
READY TEMPLATES
Drafts JSON-LD for Organization, LocalBusiness, and Article/BlogPosting using verified values or clearly marked placeholders. Human verification and validation are required before publishing.
// SCHEMA STATUS

Which schema types are currently supported?

Google regularly updates which schema types produce rich results. Claude SEO tracks these changes so you never waste time implementing deprecated markup or miss new opportunities.

Schema.org vocabulary and Google Search features

Schema.org vocabulary includes types such as Organization, LocalBusiness, SoftwareApplication, Product, Offer, Article, BreadcrumbList, WebSite, Person, VideoObject, Event, JobPosting, Course, QAPage, and DiscussionForumPosting. Google's Search Gallery supports a narrower set of documented features with its own required and recommended properties. A valid Schema.org type is not automatically eligible for a Google rich result.

Correct markup can establish eligibility, but it never guarantees display. Reject fake or misleading review data, disclose incentives where applicable, and verify that every structured value matches visible page content.

No Google Search rich result

TypeStatus
FAQPageFAQ rich results retired for all sites on May 7, 2026. Keep only when useful for non-Google or internal semantics. Use QAPage for genuine user Q&A.
DatasetStill supported by Google Dataset Search, but it does not have a Google Search rich-result surface.

Deprecated types (never recommend)

TypeStatus
HowToRich results removed September 2023
SpecialAnnouncementDeprecated July 2025
CourseInfo, EstimatedSalary, LearningVideoRetired June 2025
ClaimReviewRetired from rich results June 2025
VehicleListingRetired from rich results June 2025
Practice ProblemRich Results Test and Search Console support removed January 6, 2026
Book ActionsRemoved from Google rich results
// USAGE

How do you run a schema analysis?

Step 1: Install Claude SEO

git clone --depth 1 https://github.com/AgriciDaniel/claude-seo.git
bash claude-seo/install.sh

Step 2: Open Claude Code

claude

Step 3: Run the schema analysis

/seo schema <url>

The tool detects existing schema, checks each block, and generates draft JSON-LD for possible additions. Verify every value, fill any placeholder, and test the markup before publishing.

Output files

  • SCHEMA-REPORT.md - detection results, validation status for each schema block, and recommendations for missing schema types
  • generated-schema.json - review-ready JSON-LD drafts that require value verification and validation before publishing
// FAQ

QUESTIONS ABOUT SCHEMA MARKUP

Claude SEO detects all three Schema.org formats: JSON-LD (script type application/ld+json), Microdata (itemscope, itemprop attributes), and RDFa (typeof, property attributes). It always recommends JSON-LD as the primary format since it is Google's stated preference.
// RELATED SKILLS

EXPLORE MORE

TECHNICAL SEO

Deep technical analysis across 9 categories including crawlability, indexability, security headers, and Core Web Vitals.

TECHNICAL SEO TOOL →

FULL SITE AUDIT

The full-site audit delegates to up to 15 specialists, including structured-data review, and combines their evidence in one prioritized report.

FULL AUDIT TOOL →

GEO / AI SEARCH

Optimize for AI Overviews, ChatGPT, and Perplexity. Schema provides machine-readable entity context, but it does not guarantee AI citations.

GEO OPTIMIZATION →

E-COMMERCE SEO

Product schema is one slice. The e-commerce skill validates Product, Offer, AggregateRating, Review against Merchant Center requirements.

E-COMMERCE SEO →

SCHEMA IMAGE GENERATION

Recipe, Article, Product schema needs images. The image-gen extension produces images sized and styled to current Google requirements.

IMAGE GEN EXTENSION →

SEO DRIFT MONITORING

Schema gets stripped by security middleware, lost in refactors. Drift catches when JSON-LD blocks disappear on deploy.

DRIFT MONITORING →

VIEW ALL 25 SKILLS →

// FROM THE BLOG

RELATED GUIDE

SCHEMA MARKUP GENERATOR GUIDE

How to generate, detect, and validate JSON-LD for 30+ Schema.org types.

READ GUIDE →

VALIDATE YOUR SCHEMA
IN ONE COMMAND.

$
git clone --depth 1 https://github.com/AgriciDaniel/claude-seo.git && bash claude-seo/install.sh
VIEW ON GITHUB ALL SKILLS >