Standards the agent follows — no negotiation.
Building house institutional websites: the STRUCTURE/information architecture (page inventory — Home, About, Services index+detail, Legal index + 7 legal subpages, Contact, FAQ, Blog, Careers…), per-page content templates/checklists, the legal pages as SCAFFOLD (structure + mandatory LGPD points — NOT legal advice), structured data (JSON-LD: Organization+WebSite, BreadcrumbList, LocalBusiness, Service/Product, FAQPage, Article) and the minimal sitemap.xml+robots.txt+hreflang so the site is born indexable. Frontend ENGINEERING delegates to schematize-web; fine/multi-domain SEO to schematize-seo.
$ schematize install institutionalNon-negotiable floors
Never switched off "to go faster". Full list in the skill.
Commands
Installed in .claude/commands/. Run /institutional-help anytime.
/institutional-helplista todos os comandos do schematize-institutional/institutional-scaffolda partir de um input mínimo (nome da empresa, serviços/produtos, idiomas, tem endereço físico? é e-commerce?) gera o inventário de páginas (baseline + opcionais), os templates de conteúdo por página, os blocos JSON-LD por tipo e o sitemap.xml + robots.txt (+ hreflang se multi-idioma) — e grava no archive/institutional-loadcarrega à força TODO o corpo normativo (IA/inventário, conteúdo, legais, dados estruturados, sitemap) e passa a aplicá-lo/institutional-claudecria ou mescla o CLAUDE.md sempre-on do institucional na raiz do repo/institutional-cccontext compact: gera handoff no archive e roda /compact/institutional-handoffgera o handoff (context.md + checklist.md) sem compactarFor AI agents
Install with no human in the loop. Metadata in llms.txt.
- Run: git clone https://github.com/schematizeme/skill-institutional.git && ./skill-institutional/install.sh
- Or download: https://github.com/schematizeme/skill-institutional/releases/latest/download/skill-institutional.zip and unzip into .claude/skills/
- Copy assets/CLAUDE.md to the repo root (always-on rule).
- Confirm with /institutional-help.
토론
불러오는 중…