v1.4.0 · 3 skills · 47 commands · 77 reference docs · parallel audit sub-agents · report export · deterministic anti-pattern detector
From zero knowledge to hero website. NullToHero gives Claude a complete design, SEO and motion engineering vocabulary so anyone, even with no prior experience, can build websites that look professional, rank on Google, and pass accessibility audits.
By Marius Yvard · Apache 2.0
This repo is also a Claude Code marketplace. Two install paths:
A. As a marketplace (recommended, supports auto-update)
/plugin marketplace add MariusYvard/NullToHero
/plugin install null-to-hero@null-to-hero-marketplace
Future releases pull automatically with /plugin marketplace update null-to-hero-marketplace.
B. As a direct plugin
/plugin install MariusYvard/NullToHero
C. Manual install (Unix/macOS/Linux)
bash <(curl -fsSL https://raw.githubusercontent.com/MariusYvard/NullToHero/main/install.sh)D. Manual install (Windows PowerShell)
powershell -ExecutionPolicy Bypass -File install.ps1/seo audit now delegates each dimension to a dedicated sub-agent file in skills/seo/agents/. When the Task tool is available (Claude Code / Cowork), all 5 run in parallel — technical, content, schema, GEO, performance — then aggregate into a unified score and ACTION-PLAN. Sequential fallback when Task is unavailable.
New /seo report [url|file|generate] command formats any audit output into a client-ready deliverable: Markdown with ASCII score gauges and color-coded tables, or PDF via the Cowork PDF skill.
All /seo commands now produce action plans in a unified format: Quick Wins (< 1h), 1-Week Fixes, 1-Month Projects, Backlog. Template lives in references/action-plan.md.
tests/validate.js now checks minimum line counts per file (26 files tracked). A truncated rewrite fails the check immediately — run node tests/validate.js before every commit.
The SEO skill expands from 7 to 18 commands, the GEO command gains a weighted scoring model and two new subcommands, and the repo ships its first contributor tooling.
| Command | What it does |
|---|---|
sitemap [url|generate] |
XML sitemap validation and generation with industry templates |
images [url] |
Image SEO audit — alt text, WebP/AVIF formats, lazy loading, CLS, LCP |
local [url] |
Local SEO — Google Business Profile, NAP consistency, citations, reviews, LocalBusiness schema |
hreflang [url|generate] |
Hreflang validation and generation for multilingual/multi-region sites |
programmatic [url|plan] |
Programmatic SEO — URL patterns, quality gates (warn 100+, hard stop 500+), deduplication |
competitor-pages [url|generate] |
"X vs Y" and "alternatives to X" pages with feature matrices and schema |
cluster [keyword] |
Semantic keyword clustering by intent, gap analysis, pillar-and-cluster architecture |
sxo [url] |
Search Experience Optimization — intent alignment, page-type matching, satisfaction signals |
drift [url] baseline|compare|history |
SEO drift monitoring — capture baseline, detect changes, track history |
backlinks [url] |
Backlink profile analysis via free sources (Moz, Bing, Common Crawl, GSC) |
ecommerce [url] |
E-commerce SEO — product pages, category pages, faceted navigation, Product schema |
/seo geo now uses a weighted scoring model across 6 dimensions (citability 25 %, brand authority 20 %, content quality 20 %, technical 15 %, structured data 10 %, platform optimization 10 %) and reports platform subscores for Google AI Overviews, ChatGPT, Perplexity and Bing Copilot separately. Two new subcommands:
geo quick [url]— 60-second GEO snapshot with top 3 quick winsgeo compare [url]— delta tracking against a stored baseline
The AI crawler list expands to 14 bots: GPTBot, OAI-SearchBot, ChatGPT-User, ClaudeBot, PerplexityBot, CCBot, anthropic-ai, Bytespider, cohere-ai, Diffbot, AI2Bot, Applebot-Extended, FacebookBot, PetalBot.
CHANGELOG.md— full release historyCONTRIBUTING.md— contributor guide (reference format, quality standards, PR workflow)install.sh/install.ps1— one-line installers for Unix and Windowstests/validate.js— reference file integrity validator (node tests/validate.js)
1.2.0 — Design foundations: Gestalt principles, UX research methodology, information architecture, journey mapping, WCAG 2.2, image strategy, form patterns. Three new commands (research, ia, journey). 25 new anti-pattern rules.
1.1.0 — Parallax engineering reference and Playwright audit script. One new command (parallax). 14 new anti-pattern rules.
Build, polish, and ship production-grade interfaces.
| Command | What it does |
|---|---|
build [feature] |
Build a full UI feature end-to-end with a confirmed design brief |
plan [feature] |
Plan UX and visual direction before writing code |
research [scope] |
UX research planning, method selection, persona and journey synthesis |
ia [target] |
Information architecture, card sorting, tree testing, navigation patterns |
journey [persona] |
Generate empathy maps, journey maps, or service blueprints |
setup |
Create PRODUCT.md — brand, audience, tone, anti-references |
critique [target] |
Design review with heuristic scoring |
audit [target] |
Technical quality checks — a11y, performance, responsive, WCAG 2.2, image strategy, forms |
polish [target] |
Final quality pass before shipping |
amplify [target] |
Make bland designs bolder — typography, color, presence |
simplify [target] |
Reduce visual noise, strip to essence |
animate [target] |
Add purposeful motion and micro-interactions |
typeset [target] |
Typography audit, font selection, hierarchy |
layout [target] |
Spacing systems, visual rhythm, grid |
adapt [target] |
Mobile, tablet, desktop, print adaptation |
clarify [target] |
UX copy, error messages, button labels, empty states |
onboard [target] |
First-run flows, empty states, feature discovery |
delight [target] |
Micro-interactions, personality, satisfying feedback |
launch [target] |
Production hardening and Core Web Vitals optimization |
overdrive [target] |
View Transitions API, WebGL, scroll-driven animations |
parallax [target] |
Multi-layer depth, scrollytelling, AI-adaptive motion governance, WCAG 2.2.2 compliance |
document |
Generate DESIGN.md from existing project code |
extract [target] |
Pull reusable tokens and components into a design system |
tokens [project] |
Audit or create a two-layer CSS token system |
live [target] |
Interactive variant mode (requires running dev server) |
Get found on Google and AI search engines.
| Command | What it does |
|---|---|
audit [url] |
Full site SEO audit — crawls up to 500 pages, scores 7 dimensions, outputs ACTION-PLAN.md |
page [url] |
Deep single-page analysis — title, meta, H1-H6, schema, images, content score |
plan [business-type] |
Complete SEO strategy — architecture, content pillars, keyword plan, 4-phase roadmap |
technical [url] |
Robots.txt, sitemaps, Core Web Vitals, mobile, security headers, AI crawlers |
schema [url] |
Detect, validate, and generate Schema.org JSON-LD |
content [url] |
E-E-A-T, readability, keyword density, AI citation readiness |
geo [url] |
Weighted GEO score — Google AIO, ChatGPT, Perplexity, Bing Copilot subscores |
geo quick [url] |
60-second GEO snapshot with top 3 quick wins |
geo compare [url] |
GEO delta tracking against a stored baseline |
sitemap [url|generate] |
XML sitemap validation and generation |
images [url] |
Image SEO — alt text, formats, lazy loading, CLS, LCP |
local [url] |
Local SEO — GBP, NAP, citations, reviews, LocalBusiness schema |
hreflang [url|generate] |
Hreflang validation and generation |
programmatic [url|plan] |
Programmatic SEO with quality gates |
competitor-pages [url|generate] |
Comparison and alternatives pages with feature matrices |
cluster [keyword] |
Semantic keyword clustering and content architecture |
sxo [url] |
Search Experience Optimization — intent alignment and satisfaction signals |
drift [url] baseline|compare|history |
SEO drift monitoring |
backlinks [url] |
Backlink profile analysis |
ecommerce [url] |
E-commerce SEO — products, categories, faceted navigation |
report [url|file|generate] |
Format audit output as Markdown deliverable or PDF |
Cross-command workflows:
- New site:
plan→ build →technical→schema→sitemap→audit→report - Existing site:
audit→technical→content→geo→backlinks→report - Page not ranking:
page→content→schema→sxo - Local business:
local→schema→geo - Keyword strategy:
cluster→plan→programmatic - Before redesign:
drift baseline→ redesign →drift compare - E-commerce:
ecommerce→schema→sitemap→images - Client deliverable: any command →
report(Markdown) orreport+ PDF export
Three tools to run before every ship.
| Command | What it does |
|---|---|
detect [target] |
Deterministic anti-pattern scan — focus rings, touch targets, reduced-motion, parallax violations, WCAG 2.2, image strategy, forms |
preview [target] |
Real Chromium screenshot — desktop and mobile, reads back visually, fixes bugs in a loop |
review [file] |
Design engineering code review — motion crimes, a11y violations, Before/After table with score |
Recommended pre-ship sequence:
/inspect detect index.html
/inspect preview index.html
/inspect review index.html
NullToHero ships 75 reference documents that Claude loads on demand.
gestalt, cognitive-load, layout, typography, color-and-contrast, dark-mode-engineering, spatial-design, shape, polish, distill, bolder, quieter, amplify, simplify
ux-research, personas, journey-mapping, information-architecture, heuristics-scoring, critique, shape, teach
accessibility-engineering, wcag-2-2, form-patterns
motion-design, animation-engineering, animate, delight, parallax, overdrive, interaction-design
optimize, image-strategy, responsive-design, adapt, harden, launch
component-patterns, creative-patterns, css-architecture, design-tokens, tokens, extract, brand, document
audit, plan, technical, schema, content, page, geo
sitemap, images, local, hreflang, programmatic, competitor-pages, cluster, sxo, drift, backlinks, ecommerce
ecommerce-plan, saas, local-service, publisher, agency, generic
detect, preview, review
/siteasy research ← understand the users
/siteasy ia ← validate the structure
/siteasy setup ← define brand, audience, tone
/siteasy plan ← plan UX before coding
/seo plan ← build SEO strategy in parallel
/seo cluster ← define keyword architecture
↓
/siteasy build ← build the interface
/siteasy layout ← fix spacing and rhythm
/siteasy adapt ← make it responsive
↓
/siteasy amplify ← make it beautiful
/siteasy animate ← add motion
/siteasy parallax ← depth and scrollytelling
↓
/inspect detect ← catch anti-patterns
/inspect preview ← see it in a real browser
/inspect review ← final code quality gate
↓
/seo audit ← full SEO check
/seo technical ← fix technical blockers
/seo schema ← add structured data
/seo sitemap ← generate and validate sitemap
/seo images ← image SEO pass
/seo geo ← AI search visibility
↓
/siteasy launch ← harden for production
NullToHero works best with two files in your project root:
- PRODUCT.md — who your users are, your brand, tone, anti-references. Create with
/siteasy setup. - DESIGN.md — colors, typography, components. Generate with
/siteasy document.
- Node.js — required for
/inspect preview,/inspect detect, and the validator (tests/validate.js) - Playwright — auto-installed on first
/inspect previewrun
- 1.4.0 (May 2026) — Group C: parallel sub-agents for
/seo audit(5 agent files),/seo reportcommand (Markdown + PDF export), standardized ACTION-PLAN template across all commands, file integrity checks in validator. - 1.3.0 (May 2026) — 11 new SEO commands (sitemap, images, local, hreflang, programmatic, competitor-pages, cluster, sxo, drift, backlinks, ecommerce). GEO upgraded with weighted scoring, platform subscores,
/geo quick,/geo compare. Installer scripts, CHANGELOG, CONTRIBUTING, reference validator. - 1.2.0 (May 2026) — Design foundations: Gestalt, UX research, IA, journey mapping, WCAG 2.2, image strategy, form patterns. Three new commands (research, ia, journey). 25 new anti-pattern rules.
- 1.1.0 (May 2026) — Parallax engineering reference and audit script. One new command (parallax). 14 new anti-pattern rules.
- 1.0.0 (April 2026) — Initial release: siteasy, seo, inspect with 31 commands and core references.
See CONTRIBUTING.md for how to add reference files, new commands, or bug fixes.
node tests/validate.js # run before opening a PRApache 2.0 — github.com/MariusYvard/NullToHero