Schema Markup Assistant
Detect, validate, and generate SEO-ready Schema.org structured data.
The skill requires truthful, verifiable data and defines handling for unreachable URLs, missing markup, invalid JSON-LD, and deprecated types. It does not specify least privilege, user confirmation, data flows, sensitive-data or credential handling, external-request risk, or rollback, so points were deducted.
The detection, validation, and generation workflow is broadly consistent and covers several abnormal-input cases. However, dependencies are not fully documented in the target file, referenced schema-types.md and templates.json are not supplied, and there are no path-specific tests, parsing commands, or reproducible diagnostic failure procedures, so points were deducted.
Trigger terms, URL input, and schema-analysis scenarios are clear, and output artifacts are named. Non-fit boundaries, Chinese-language support, mainland-China reachability, and constraints for varied page or schema complexity are not sufficiently defined, so points were deducted.
The file includes name, description, version, MIT license, structured sections, templates, output formats, and an error-handling table. Installation and dependency notes, maintenance ownership, update path, changelog, complete reference coverage, and FAQs are incomplete, so points were deducted.
The workflow covers detection, basic validation, recommendations, and JSON-LD generation, with Markdown and JSON outputs. Static evidence does not prove that generated results are directly production-ready, validation coverage is limited, and comparative benefit over manual or other tools is not demonstrated, so points were deducted.
The supplied materials include some dated, traceable Google-related references and require truthful verifiable data. The target skill lacks direct evidence links, dedicated tests, or independent reproduction records, and several time-sensitive claims cannot be verified from the target file alone, so points were deducted.
- The referenced references/schema-types.md and schema/templates.json were not supplied, so the stated type status and template availability cannot be statically verified.
- The skill processes user-provided URLs and may cause external requests, but the target file does not describe privacy, SSRF protections, credential boundaries, or confirmation requirements.
- The May 2026 status claims and Google policy changes are time-sensitive; this review relied only on the supplied material and performed no execution or independent verification.
- The templates contain placeholders; all entities, contact details, dates, and URLs require human verification before use.
What does this skill do, and when should you use it?
This is the standalone Schema markup skill inside the Claude SEO repository. It detects, validates, and generates Schema.org structured data, with JSON-LD as the preferred format while also checking Microdata and RDFa. It checks required properties, common data errors, Google rich result support, and deprecated types. Its stated outputs are a validation report and ready-to-use JSON-LD snippets.
Reads page source and scans for JSON-LD, Microdata, and RDFa; checks @context, @type, data types, placeholder text, absolute URLs, and date formats; validates required properties by schema type; checks Google-supported rich result types; flags deprecated types; identifies the page type and generates JSON-LD with required and recommended properties; reports URL access failures, missing markup, and invalid JSON-LD syntax with corrective guidance; produces SCHEMA-REPORT.md and generated-schema.json.
- An SEO consultant reviews a client page for missing required properties and invalid existing JSON-LD.
- A web developer generates deployable Organization, LocalBusiness, Article, or BlogPosting markup.
- A content team checks whether an article page uses an appropriate Schema.org type.
- A site owner audits structured data after a template migration for URL, date, and type errors.
What are this skill's strengths and limitations?
- Checks JSON-LD, Microdata, and RDFa.
- Combines detection, validation, error identification, deprecated-type guidance, and JSON-LD generation.
- Requires generated data to be truthful and verifiable, with explicit handling for common failure cases.
- Includes practical Organization, LocalBusiness, and Article/BlogPosting templates.
- Only some templates are shown; the full type list and templates are referenced in references/schema-types.md and schema/templates.json, which were not supplied here.
- Analysis depends on an accessible page URL; authenticated pages may not be analyzable.
- The skill does not document an independent test suite or cross-platform runtime verification.
- Its rich-result status rules follow the dates stated in SKILL.md and may require maintenance as search guidelines change.
How do you install this skill?
The skill is located at skills/seo-schema/SKILL.md in the repository. To install the public repository, use: git clone --depth 1 https://github.com/AgriciDaniel/claude-seo.git, then run bash claude-seo/install.sh. The supplied source does not document a separate installation procedure for this sub-skill.
How do you use this skill?
In Claude Code with Claude SEO installed, run /seo schema <url>, for example: /seo schema https://example.com. The skill description also supports a URL argument and is intended to activate when the user mentions schema, structured data, rich results, JSON-LD, or markup.