What does this skill do, and when should you use it?
Blog Schema is a focused Agent Skill in the claude-blog repository for generating JSON-LD structured data from blog posts. It reads article content, extracts metadata such as authorship, dates, images, FAQs, categories, and slugs, then links the resulting entities through an @graph. The output can include BlogPosting, Person, Organization, BreadcrumbList, FAQPage, ImageObject, and VideoObject. It is a good fit for teams standardizing blog markup, but it is not a general writing, keyword research, or site-wide SEO skill.
Reads a blog post and extracts the title, author details, publication and modification dates, meta description, FAQ pairs, cover and inline images, organization information, approximate word count, tags or categories, and slug. It generates BlogPosting, Person, Organization, BreadcrumbList, FAQPage, and primary ImageObject entities, plus VideoObject entities for embedded YouTube videos when metadata is available. It combines the results into one JSON-LD script using @context and @graph, stable @id references, and entity linking. Validation checks cover date ordering, headline and description length, absolute URLs, positive image dimensions, sequential breadcrumbs, minimum FAQ count, unresolved references, and deprecated schema types.
- A blogger is publishing a new article and wants a ready-to-embed JSON-LD block generated from the article file.
- An SEO or editorial team is auditing an existing post's article, author, publisher, breadcrumb, FAQ, and image markup.
- A content team wants FAQ content represented in a structure that search engines and AI citation systems can extract more consistently.
- A publishing team embeds YouTube videos and wants corresponding VideoObject markup alongside the article schema.
- A developer maintains many blog pages and needs stable @id references connecting articles, authors, publishers, and media entities.
How do you install this skill?
- The skill does not clearly require confirmation before overwriting a blog file; use a separate output file or retain a backup in practice.
- Claims about validating Google requirements, requiring at least two FAQ items, and approximately 13% higher AI-citation likelihood lack skill-specific verifiable evidence in the supplied material.
- Dates, image dimensions, author social links, and video statistics may be missing or unreliable; generated output needs human review and validation with an actual structured-data tool.
- Fallback behavior is unspecified for Chinese content, different CMSs, or inaccessible external images and YouTube metadata.
- Network access
- Local filesystem
Install it as part of the AgriciDaniel/claude-blog collection. In Claude Code 1.0.33+, run /plugin marketplace add AgriciDaniel/claude-blog, then /plugin install claude-blog@agricidaniel-claude-blog. The skill file is skills/blog-schema/SKILL.md. The README does not document a separate installation command for this sub-skill.
tmp="$(mktemp -d)"
git clone --depth 1 https://github.com/AgriciDaniel/claude-blog.git "$tmp"
mkdir -p ~/.claude/skills
cp -R "$tmp/skills/blog-schema" ~/.claude/skills/
rm -rf "$tmp"Generated from the source repository and skill path; it copies only this skill's folder. If the author's install steps above differ, follow those first. To scope it to one project, replace ~/.claude/skills with that project's .claude/skills.
How do you use this skill?
In Claude Code, run /blog schema <file> for a blog post, or ask for “schema,” “blog schema,” “json-ld,” “structured data,” or “schema markup.” The skill reads the specified article and generates the markup. It supports embedded HTML, standalone JSON, and MDX component output; whether to save the result into the post or a separate schema file is left to the user's preference.
What are this skill's strengths and limitations?
- Covers the common article, author, publisher, breadcrumb, FAQ, and image entities used by blog pages.
- Uses the @graph pattern and stable @id references for maintainable entity linking.
- Includes validation rules for Google-oriented properties and warnings for deprecated types.
- Supports embedded HTML, standalone JSON, and MDX component output.
- Its scope is structured data generation, not full blog writing, keyword research, or site-wide SEO auditing.
- FAQ validation requires at least two questions, so an article without sufficient FAQ content cannot pass that check.
- The video workflow is centered on YouTube, with metadata dependent on embed fallbacks or available blog-google data.
- The SKILL.md provides no standalone test suite, CMS-specific setup procedure, or separate installation workflow.
- The README's claim of approximately 13% higher AI citation likelihood is not accompanied by a verification method in the supplied skill material.
How does this skill compare with similar options?
Side by side with related skills; every score comes from the same FSRS standard.
| Skill | FS score | Stars | Last updated | License |
|---|---|---|---|---|
| Blog Schema: JSON-LD Markup Generator this page | 46 · Use with care | ★ 2.4k | 14d ago | MIT |
| Schema Markup Assistant | 52 · Use with care | ★ 19k | 23d ago | MIT |
| SERP Markup Builder | 58 · Recommended | ★ 3k | 13d ago | Apache-2.0 |
| Blog Locale Audit | 50 · Use with care | ★ 2.4k | 14d ago | MIT |
| Blog Translate: SEO Blog Localization | 48 · Use with care | ★ 2.4k | 14d ago | MIT |
The README lists an optional /seo-schema companion skill for deeper schema validation and generation. Blog Schema is narrower: it focuses on generating a complete JSON-LD @graph from a blog post.
How did FollowSkills review this skill?
The skill is limited to generating JSON-LD and declares no tool, credential, or external-write capability; no malicious behavior is shown. However, it permits saving output to the blog file at the user's preference without explicit write confirmation, backup, or rollback guidance, and it does not disclose data flow involving user content and external URLs, so points are deducted.
The workflow and examples cover the main happy path, with a defined validation checklist for fields and references. However, there are no skill-specific generation tests, abnormal-input procedures, or controlled failure messages; “validation” is specified textually rather than provided as reproducible implementation, so points are deducted.
Trigger phrases, file-path input, and output choices are reasonably clear for blog structured-data generation. Boundaries for unsupported cases, missing fields, Chinese-language content, and CMS-specific behavior are not defined, while FAQ and video metadata require additional information, so points are deducted.
The file has standard frontmatter, an MIT license declaration, progressive workflow sections, examples, and some limitation notes. It lacks its own version, changelog, maintenance owner, or update path, and installation/dependency guidance is mostly repository-level rather than skill-local, so points are deducted.
The templates can produce directly embeddable JSON-LD with linked entities and cover the core task. However, correctness depends on manual completion of site, author, image, and video metadata; FAQ rules and AI-citation benefits are unsupported in the supplied material, and save behavior and CMS compatibility are unclear, so points are deducted.
The repository includes general CI and test evidence, but the supplied files contain no dedicated reproducible tests for blog-schema generation, reference resolution, or Google validation, and no third-party execution evidence. Static review therefore supports only limited auditability.
Open a dimension to read why it scored that way
Evidence confidence：Low — Mostly static review, author material or a limited demo; useful for discovery, not high-risk decisions.
See the full review method →