菜单
Generate JSON-LD structured data markup for any website. Supports HowTo, FAQPage, Article, BlogPosting, Organization, WebSite, BreadcrumbList, Product, LocalBusiness, and MedicalWebPage types.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Optimize a content draft or existing page for AI citability using the CITE framework (Clear answer, Information density, Taxonomic structure, Expert authority).
Run a comprehensive GEO visibility audit on any website. Analyzes AI citability, structured data, AI crawler access, llms.txt, and content quality. Produces a 0-30 citability score and prioritized action plan.
MetaBot HTTP API for agent collaboration: talk to other bots, schedule tasks, manage bots and peers. Use when the user wants to delegate work to another bot, schedule tasks, create/remove bots, or check peer status.
Read and write shared memory documents. Use this when you need to save knowledge, notes, research findings, or project context for future reference across sessions. Also use it to look up previously stored information.
The meta-skill: create AI agent teams, individual agents, or custom skills for any project. Use when the user wants to generate a complete .claude/ agent team, create a single agent, or create a single skill.
| name | schema-generator |
| description | Generate JSON-LD structured data markup for any website. Supports HowTo, FAQPage, Article, BlogPosting, Organization, WebSite, BreadcrumbList, Product, LocalBusiness, and MedicalWebPage types. |
| user-invocable | true |
| allowed-tools | Bash, Read, Grep, Glob, WebFetch |
| context | fork |
You generate production-ready JSON-LD structured data for any website to maximize AI engine comprehension and rich result eligibility.
The user may provide:
If a brand.md exists in the project root, read it for publisher details (name, url, logo). Otherwise, infer from the target page.
| Page Type | Schema Types to Generate |
|---|---|
| Step-by-step guide / tutorial | HowTo + FAQPage + BreadcrumbList |
| Blog post / article | BlogPosting + FAQPage + BreadcrumbList |
| Long-form research article | Article + FAQPage + BreadcrumbList |
| Category / collection page | CollectionPage + ItemList + BreadcrumbList |
| Homepage | WebSite + Organization |
| About / company page | AboutPage + Organization |
| Product page | Product + FAQPage + BreadcrumbList |
| Local business | LocalBusiness + FAQPage |
| Health / medical content | MedicalWebPage + FAQPage + BreadcrumbList |
python3 scripts/check-schema.py <url>
Review the output: what schema types are already present, what's missing, what's incomplete. Use this as the baseline — only generate what's needed.
Read the page/content and identify:
Produce each schema block as a separate <script> tag:
<script type="application/ld+json">
{
"@context": "https://schema.org",
"@type": "...",
...
}
</script>
# Validate JSON syntax
echo '<json-ld-content>' | python3 -m json.tool > /dev/null && echo "VALID" || echo "INVALID"
Ensure consistency:
For each page, produce:
<head>)Save to schema/[page-slug]-schema.json.