Digital Identity Architect
So Google and AIs understand your website unambiguously — real entities, a clean knowledge graph, zero schema spam.
SKILL.md in the open agentskills.io standard — works directly in Claude, ChatGPT/Codex, Cursor, Copilot & more.
Digital Identity Architect is an AI skill for a website's digital identity — structured data, Schema.org/JSON-LD and the knowledge graph. It first understands a site's real entities, models an entity graph with stable @ids, then implements the markup framework-correctly and validates it — honestly, with no rich-result hunting. Not about digital identity in the eID/login sense.
What this skill does
- Analyzes the website and/or codebase and captures real entities with evidence and confidence levels
- Surfaces duplicate/conflicting schema (theme vs plugin vs app) and decides one owner per type incl. disable steps
- Models an entity graph with stable @ids (industry model), an entity home and honest sameAs
- Rates rich results in two columns — SERP eligibility (dated, verified live) vs entity/AI value
- States the platform split honestly (Google/Bing verifiably use markup, ChatGPT & co. read visible text)
- Implements after approval, framework-correctly and XSS-safe (Next.js, Astro, Nuxt, SvelteKit, WordPress, Shopify …)
- Validates twice plus its own third check (@id wiring, content parity, cross-page consistency) and delivers a prioritized backlog
Description
Digital Identity Architect is a skill for the digital identity of websites — it makes search engines and AI systems understand unambiguously WHO and WHAT a website is. Instead of filling schema templates, it works like an architect: it first analyzes the real entities (organization, brands, products, software, people, content) plus the existing markup, surfaces the documented no. 1 real-world problem — **duplicate, conflicting schema from theme, plugin and shop app** — and decides one owner per type. From that it models an **entity graph** with stable @ids after the industry model (full node set per page, referenced within the page), including an entity home, legal identifiers and honest sameAs. Before any implementation stands a **strategy gate** — it implements only after your approval, then framework-correctly and XSS-safe (Next.js, Astro, Nuxt, SvelteKit, WordPress, Shopify, Hugo, headless CMS …). Its foundation is **evidence-based honesty**: rich-result advice comes from a dated two-column matrix (SERP eligibility, verified live against Google's changelog, vs stable entity/AI value — FAQ stars have been history since 05/2026), and the platform split is stated openly — Google and Bing verifiably use markup, ChatGPT & co. primarily read visible text. Add the SSR gate (client-injected JSON-LD is invisible to AI crawlers), content parity as a hard rule, Wikidata guardrails (never auto-created) and a saveable Identity Profile for consistent follow-up runs. What is not real is not marked up. What is not proven is not promised.
Examples
What it does not do
- No schema spam and no rich-result hunting — it models only the real identity
- Invents no entities, ratings, figures or effect claims (self-serving stars are refused)
- Not an eID/login/citizen-ID topic — that's what the Government Navigator is for
- No breadth audit for crawling/performance/llms.txt (→ Web Vitals / Web AI Vitals) and no content strategy (→ SEO Briefing)
- Never implements before the strategy gate, never changes files unasked, never auto-creates Wikidata items
Compatibility & tech
- Tested (internal)
- 5 scenarios
- Recommended runtime
- Claude Opus or Sonnet; with web/shell access for live analysis and verify-live — without access the skill works from supplied HTML/code and states the data basis
- Modes
- Audit (check only) · Fill gaps only · Modernize · Full (with strategy gate) · Compare live vs codebase
- Inputs
- Website URL · HTML / existing JSON-LD · project folder / codebase · existing Identity Profile
- Output format
- Identity report (coverage + duplicate audit + findings) + entity graph (Mermaid) + prioritized backlog + code/snippets + Identity Profile
- Subcategory
- Structured data & knowledge graph
- License
- Proprietär
Security profile
Runs entirely on your machine with your own AI — no external runtime, no running costs.
Only instructions, templates and references — no executable scripts.
Accesses external sources / the network in normal use (e.g. live pages, search/data APIs).
What you get
- digital-identity-architect-1.0.0/7 files
- .claude-plugin/marketplace.json
plugins/digital-identity-architect/skills/digital-identity-architect/7 files
- SKILL.md
- manifest.json
references/5 files
- digital-identity-architect-ai-evidence.md
- digital-identity-architect-checklist.md
- digital-identity-architect-entity-modeling.md
- digital-identity-architect-implementation.md
- digital-identity-architect-rich-results.md
- .agents/skills/digital-identity-architect/→ universal — same content (Codex, Cursor, Copilot, Gemini, Windsurf, Cline)
- LICENSE.txt
Installation
Also works as a chat prompt
No AI tool? Paste it into Claude, ChatGPT or Gemini and use the method right away.
This skill ships no scripts, so the prompt carries its full method.
Works best when your chat has web access.
Installing is the full version — it triggers automatically, runs its scripts and loads references as needed. As a chat prompt you drive the method by hand.
Unlock to copy the ready-to-paste prompt — then in “My Skills”.
Reviews
No reviews yet – be the first.
Note
Best with the URL or project folder of the real website. The skill implements only after your approval (strategy gate), performs no file changes unasked and never auto-creates Wikidata entries. Rich-result rules change constantly — with web access the skill verifies them live against Google's official docs and cites the retrieval date; a specific SERP display or AI citation is not guaranteed.
Changelog
- v1.0.010.07.2026Initial release: 5-step workflow with a strategy gate (never implements unasked), duplicate-schema audit with one owner per type, @id design system after the industry model, entity home & knowledge-panel reality, Wikidata guardrails, a dated two-column rich-results matrix (SERP eligibility vs entity/AI value) incl. verify-live, an evidence-based platform split (Google/Bing verifiably use markup, ChatGPT & co. read visible text), SSR gate, XSS-safe framework patterns (Next.js/Astro/Nuxt/SvelteKit/WordPress/Shopify and more) and an Identity Profile. Five references incl. the evidence file and a failure catalog.
Frequently asked questions
What does Digital Identity Architect do?
Digital Identity Architect is an AI skill for a website's digital identity — structured data, Schema.org/JSON-LD and the knowledge graph. It first understands a site's real entities, models an entity graph with stable @ids, then implements the markup framework-correctly and validates it — honestly, with no rich-result hunting. Not about digital identity in the eID/login sense.
Which structured data does my website actually need?
Digital Identity Architect answers exactly that — it first analyzes your site's real entities (organization, products, people, content), finds gaps and duplicates, and recommends only markup that reflects your true identity instead of schema spam for retired rich results.
Is this the same as digital identity or eID for logging in?
No — Digital Identity Architect handles your website's identity for search engines and AI systems (Schema.org/JSON-LD, knowledge graph). For digital identity in the government sense (eID, eIDAS, logins), skills-for-ai has the Government Navigator.
Does schema markup get my site into ChatGPT answers?
Honestly, not directly — ChatGPT, Claude and Perplexity primarily read visible text. Verifiably, Google (Knowledge Graph, AI Mode fan-outs) and Bing/Copilot do use structured data. Digital Identity Architect tells you per platform what really works instead of making blanket promises.
Is FAQ schema still worth it?
As a Google rich result, FAQ was fully retired in May 2026 — as a content pattern, real questions and answers stay valuable for AI understanding. Digital Identity Architect rates every type honestly in two columns (SERP eligibility, verified live, vs entity/AI value).
Can the skill add stars and ratings?
Only from real user reviews visible on the page — self-serving ratings about your own organization are excluded by Google policy, and Digital Identity Architect refuses invented stars on principle.
How is it different from Web AI Vitals?
Web AI Vitals audits AI visibility in breadth (7 categories); Digital Identity Architect is the depth architect for exactly one of them — it models and implements the entity graph with Schema.org/JSON-LD. Ideal used back to back.
Which AI tools does Digital Identity Architect work with?
Claude · ChatGPT/Codex · Cursor · Copilot · Gemini CLI · Windsurf · Cline
How do I use Digital Identity Architect?
Digital Identity Architect is a SKILL.md in the open agentskills.io standard: install it with one command (npx) or download it and add it to your AI tool — Claude (Projects), ChatGPT (Custom GPT), Cursor, Copilot, Gemini CLI and more. No code needed.