olsonbd/superpowers-antigravity

implementing-schema-org

Use when defining data models, creating Payload Collections, or structuring API responses - ensures strict compliance with Schema.org standards for interoperability and SEO, preventing ad-hoc naming and flattened structures

ソースを見る

使い方

  1. インストールコマンドをコピーし、自分の環境へ導入する前にソースを確認します。
  2. 必要なツール、認証情報、対応環境を原文で確認します。
  3. 入力と成果物を指定し、出力を資料と照合します。
npx skills add https://github.com/olsonbd/superpowers-antigravity --skill implementing-schema-org

開始前の確認

  • 導入にはモデル契約、API枠、ファイルへのアクセス許可は含まれません。
  • スター数や導入数は活動の指標であり、独立した品質・安全性試験ではありません。
リポジトリの原文

見出し、例、コード、表、リンク、参照画像を含む原文を表示しています。

Implementing Schema.org Data Models

Overview

Schema.org is the internet's standard for structured data. Adopting it prevents "bikeshedding" over field names and ensures our data is ready for SEO, AI agents, and external integrations (like calendars) from day one.

Core Principle: If Schema.org has a standard for it, use it. Do not invent your own field names.

When to Use

  • Creating a new Collection (e.g., Venues, Events, Users)
  • Defining API responses
  • Mapping data for export

The Rules

  1. Strict Naming: Use the exact property name from Schema.org (e.g., postalCode not zip, givenName not firstName).
  2. Preserve Structure: If Schema.org defines a property as a nested Type (like address -> PostalAddress), you MUST nest it. Do not flatten it for convenience.
  3. Use Standard Types: Map your entities to the most specific Schema.org Type available (e.g., Event -> SocialEvent if applicable).

Common Mappings

Our ConceptSchema.org TypeKey Fields
VenuePlacename, address (PostalAddress), geo (GeoCoordinates), publicAccess
EventEventname, startDate, endDate, location (Place), organizer (Person/Organization)
UserPersongivenName, familyName, email, telephone

Implementation Pattern (Payload CMS)

❌ BAD: Flattened & Ad-Hoc

typescript
// Venues.ts
fields: [
  { name: "street", type: "text" }, // Violation: Flattened
  { name: "zip", type: "text" }, // Violation: Wrong name
  { name: "lat", type: "number" }, // Violation: Wrong name & flattened
];

✅ GOOD: Nested & Standard

typescript
// Venues.ts
fields: [
  {
    name: "address", // Schema.org property
    type: "group", // Preserves structure
    fields: [
      { name: "streetAddress", type: "text" },
      { name: "addressLocality", type: "text" }, // City
      { name: "addressRegion", type: "text" }, // State
      { name: "postalCode", type: "text" }, // Zip
      { name: "addressCountry", type: "text" },
    ],
  },
  {
    name: "geo",
    type: "group",
    fields: [
      { name: "latitude", type: "number" },
      { name: "longitude", type: "number" },
    ],
  },
];

JSON-LD Serialization

When exposing this data via API, simply wrap it to produce valid JSON-LD:

typescript
// Example response structure
{
  "@context": "https://schema.org",
  "@type": "Place",
  "name": venue.name,
  "address": {
    "@type": "PostalAddress",
    ...venue.address // Spread the nested group directly!
  },
  "geo": {
    "@type": "GeoCoordinates",
    ...venue.geo
  }
}

Benefit: Because we preserved the structure in the database, serialization is just a spread operator (...venue.address). No complex mapping code required.

Red Flags - STOP and Fix

  • "I'll just call it zip because it's shorter." -> STOP. Use postalCode.
  • "Nested groups are annoying to query." -> STOP. Structure matters more than convenience.
  • "I'll add a customField." -> Check Schema.org first. Is there really no standard field?

References

同じリポジトリから

関連する Skills

すべての Skills