사용 방법
- 설치 명령을 복사하고 자신의 에이전트 환경에 설치하기 전에 소스 패키지를 확인하세요.
- 필요한 도구, 인증 정보, 지원 환경은 원본 문서에서 확인하세요.
- 입력과 원하는 산출물을 설명하고 출처와 결과를 대조하세요.
npx skills add https://github.com/sergekostenchuk/ui-ux-agent-skill-system --skill technical-seo-schema-engineer시작 전 확인
- 설치에는 모델 구독, API 한도, 파일 접근 권한이 포함되지 않습니다.
- 별점 수와 설치 수는 활동 지표이며 독립적인 품질·보안 검증이 아닙니다.
원본 Skill 문서
원본 저장소의 제목, 예시, 코드, 표, 링크, 이미지를 유지해 표시합니다.
Technical SEO Schema Engineer
Use this skill after information-architecture-seo has defined page roles and before live regression validation.
Read references/schema-patterns.md before designing full metadata or JSON-LD templates.
Owns
- title and meta description rules;
- canonical and hreflang tag templates;
- JSON-LD schema patterns;
- OpenGraph and Twitter card metadata;
- sitemap, news-sitemap, RSS, robots.txt, and llms.txt output policy;
- publisher logo and ImageObject fallback policy;
- page-type metadata checklist.
Does Not Own
- semantic core;
- final URL architecture;
- internal link graph;
- live audit execution;
- hidden content;
- visual page layout.
Workflow
- Read
url-map.yaml, page roles, content model, language groups, and visible content requirements. - Select metadata and schema by page role.
- Define title/meta fallbacks and length guidance.
- Define canonical and hreflang tags from URL map.
- Define JSON-LD nodes that match visible page content.
- Define OG/Twitter image policy: media, video poster, publisher card, or fallback OG card.
- Define sitemap/RSS/robots/llms.txt inclusion rules.
- Produce schema/head guidance using assets/jsonld-templates.md and assets/head-metadata.template.html.
- Hand off validation to
seo-regression-validator.
Non-Negotiables
- Schema must match visible content.
- Do not add FAQPage unless the FAQ block is visible to users.
- Do not add review, rating, price, author, publisher, date, image, or source claims that are absent or unsupported.
- Do not use JSON-LD to hide extra keywords, extra entities, or alternate facts.
- NewsArticle.image must resolve to media, poster, or a stable fallback ImageObject.
- Canonical and hreflang must follow URL architecture.
- Current Google/schema.org/crawler rules require primary-source verification before production changes.
Safety And Privacy Boundaries
- Do not expose admin, API, private dashboards, raw analytics, raw IP logs, secret files, tokens, unpublished drafts, or staging URLs in metadata, schema, sitemaps, RSS, robots, or llms.txt.
- Do not weaken authentication, access control, CSP, CORS, cookies, rate limits, or private-content boundaries for crawlability.
- Do not place private instructions or internal planning notes in public LLM-facing files.
- Do not treat robots.txt as a secrecy mechanism for sensitive content.
- Security, privacy, legal accuracy, accessibility, and visible user comprehension override SEO or LLM visibility tactics.
Validation
Before marking work complete:
- every page type has title, description, canonical, OG, Twitter, and schema guidance;
- every schema type has required visible-content fields;
- every news/article page has image fallback policy;
- every hreflang group is based on URL map equivalents;
- sitemap and news-sitemap rules include only intended public canonical URLs;
- robots and llms.txt rules avoid private/admin/API surfaces;
- hidden schema and duplicate canonical patterns are rejected.
Validate skill edits with:
bash
python3 $CODEX_HOME/skills/senior-skill-architect/scripts/lint_production_skill.py $HOME/SKILL/skills/technical-seo-schema-engineer
python3 $HOME/SKILL/plans/seo-llm-skill-cluster/scripts/lint_skill_cluster.py $HOME/SKILLForward tests live in evals.json.
Output Shape
Return:
- Page-type metadata matrix.
- Schema nodes and required visible fields.
- Canonical/hreflang rules.
- Sitemap/RSS/robots/llms.txt inclusion policy.
- Validation checklist and handoff to regression validator.
