bmad-plus 0.7.4 → 0.8.0

package/src/bmad-plus/packs/pack-seo/ref/cwv-thresholds.md

@@ -0,0 +1,87 @@
+# Core Web Vitals — Thresholds & Optimization Guide (March 2026)
+
+> Author: Laurent Rochetta | BMAD+ SEO Engine v2.0
+
+## Current Metrics
+
+| Metric | Good | Needs Improvement | Poor |
+|--------|------|-------------------|------|
+| **LCP** (Largest Contentful Paint) | ≤2.5s | 2.5–4.0s | >4.0s |
+| **INP** (Interaction to Next Paint) | ≤200ms | 200–500ms | >500ms |
+| **CLS** (Cumulative Layout Shift) | ≤0.1 | 0.1–0.25 | >0.25 |
+
+> **INP replaced FID on March 12, 2024.** FID was fully removed from all Chrome tools on September 9, 2024. Never reference FID.
+
+## Key Facts
+- Evaluation uses the **75th percentile** of real user data (CrUX field data)
+- Assessment at **page level** AND **origin level**
+- CWV is a **tiebreaker** signal — matters most when content quality is similar
+- Thresholds **unchanged since original definitions**
+- December 2025 core update appeared to weight **mobile CWV more heavily**
+- As of October 2025: 57.1% desktop, 49.7% mobile sites pass all three CWV
+- **Mobile-first indexing 100% complete** since July 5, 2024
+
+## LCP Diagnostic Subparts (Feb 2025 CrUX)
+
+| Subpart | Description | Target |
+|---------|-------------|--------|
+| TTFB | Server response time | <800ms |
+| Resource Load Delay | TTFB to resource request | Minimize |
+| Resource Load Time | LCP resource download | Size-dependent |
+| Element Render Delay | Loaded to painted | Minimize |
+
+**Total LCP = TTFB + Load Delay + Load Time + Render Delay**
+
+## Common Bottlenecks
+
+### LCP
+- Unoptimized hero images → use WebP/AVIF, add `<link rel="preload">`
+- Render-blocking CSS/JS → defer, async, critical CSS inline
+- Slow TTFB (>200ms) → CDN, caching, edge compute
+- Third-party blocking → defer analytics, chat widgets
+- Font delay → `font-display: swap` + preload
+
+### INP
+- Long JS tasks on main thread → break tasks <50ms, use `scheduler.yield()`
+- Heavy event handlers → debounce, requestAnimationFrame
+- Excessive DOM (>1,500 elements)
+- Third-party scripts hijacking main thread
+- Layout thrashing (forced reflows)
+
+### CLS
+- Images/iframes without width/height dimensions
+- Dynamic content injected above existing content
+- Font swap causing layout shift → preload + font-display
+- Ads/embeds without reserved space
+
+## Measurement Sources
+
+### Field Data (used for ranking)
+- Chrome User Experience Report (CrUX)
+- PageSpeed Insights (uses CrUX)
+- Search Console Core Web Vitals report
+
+### Lab Data (for debugging)
+- Lighthouse 13.0 (Oct 2025, restructured audits)
+- WebPageTest
+- Chrome DevTools
+
+## Optimization Priority
+1. **LCP** — Most impactful for perceived performance
+2. **CLS** — Most common UX issue
+3. **INP** — Critical for interactive applications
+
+## API Tools
+```bash
+# PageSpeed Insights API
+curl "https://www.googleapis.com/pagespeedonline/v5/runPagespeed?url=URL&key=API_KEY"
+
+# Lighthouse CLI
+npx lighthouse URL --output json --output-path report.json
+```
+
+## 2025–2026 Updates
+- **Lighthouse 13.0** (Oct 2025): reorganized performance categories
+- **CrUX Vis** replaced CrUX Dashboard (Nov 2025) → cruxvis.withgoogle.com
+- **LCP subparts** in CrUX (Feb 2025)
+- **Soft Navigations API** (Chrome 139+, July 2025): experimental SPA CWV measurement — no ranking impact yet

package/src/bmad-plus/packs/pack-seo/ref/eeat-criteria.md

@@ -0,0 +1,123 @@
+# E-E-A-T Evaluation Criteria (March 2026)
+
+> Author: Laurent Rochetta | BMAD+ SEO Engine v2.0
+> Based on: Google Quality Rater Guidelines (September 2025 update)
+
+## Overview
+
+E-E-A-T = **Experience, Expertise, Authoritativeness, Trustworthiness**
+
+Google's quality raters evaluate pages on these four dimensions. While not a direct ranking factor, E-E-A-T signals influence how Google's algorithms assess content quality and YMYL (Your Money or Your Life) pages.
+
+> Since March 2024, the Helpful Content System is merged into Google's core algorithm. Enforcement is continuous, not via separate HCU updates.
+
+---
+
+## Experience (25 points)
+
+First-hand, real-world experience with the topic.
+
+| Signal | Points | How to Detect |
+|--------|--------|---------------|
+| Original research or case studies | 8 | Unique data, before/after results, proprietary findings |
+| First-hand process documentation | 6 | Step-by-step from personal experience, production screenshots |
+| Unique media from direct experience | 6 | Original photos, self-recorded videos, personal screenshots |
+| Specific examples and anecdotes | 5 | Named examples, concrete details, real scenarios |
+
+**Red flags**: Generic advice that could come from any source, stock photos only, no personal perspective.
+
+---
+
+## Expertise (25 points)
+
+Demonstrable knowledge and skill in the subject area.
+
+| Signal | Points | How to Detect |
+|--------|--------|---------------|
+| Author credentials visible | 7 | Bio with qualifications, certifications, professional background |
+| Technical depth appropriate | 7 | Correct terminology, nuanced explanations, awareness of edge cases |
+| Well-sourced claims | 6 | Citations to studies, official documentation, linked sources |
+| Comprehensive topic coverage | 5 | Covers subtopics, addresses counterarguments, FAQ sections |
+
+**Red flags**: No author identified, surface-level coverage, factual errors, outdated information.
+
+---
+
+## Authoritativeness (25 points)
+
+Recognition as a trusted source by peers and industry.
+
+| Signal | Points | How to Detect |
+|--------|--------|---------------|
+| External citations / references | 7 | Backlinks from authoritative sites, cited in industry publications |
+| Brand recognition signals | 7 | Awards, partnerships, media mentions, client logos |
+| Author published elsewhere | 6 | Guest posts, conference talks, books, academic papers |
+| Expert endorsements | 5 | Testimonials from peers, quoted by other experts |
+
+**Red flags**: Brand-new domain with no web presence, no external references, no social proof.
+
+---
+
+## Trustworthiness (25 points)
+
+Transparency, accuracy, and confidence in the source.
+
+| Signal | Points | How to Detect |
+|--------|--------|---------------|
+| Contact info and physical address | 7 | Phone, email, postal address, About page |
+| Privacy policy and terms | 5 | Legal pages present, accessible, and up-to-date |
+| HTTPS and security signals | 5 | Valid SSL, HSTS, security headers present |
+| Transparent authorship and dates | 5 | Author byline, publication date, last updated date |
+| Customer proof | 3 | Verified reviews, testimonials, case studies |
+
+**Red flags**: No contact information, anonymous content, no dates, misleading claims, HTTP only.
+
+---
+
+## YMYL Pages — Higher Standards
+
+YMYL (Your Money or Your Life) topics require **elevated E-E-A-T**:
+
+| YMYL Category | Examples | E-E-A-T Bar |
+|---------------|----------|-------------|
+| Health | Medical advice, symptoms, treatments | Very High — requires medical expertise |
+| Finance | Investment, tax, insurance, banking | Very High — requires financial credentials |
+| Safety | Product safety, emergency info | Very High |
+| Legal | Legal rights, contracts, regulations | Very High — requires legal expertise |
+| News | Current events, political coverage | High — requires journalistic standards |
+| Shopping | Product reviews, comparisons | Moderate-High |
+
+---
+
+## AI Content Assessment (Sept 2025 QRG Addition)
+
+Google raters now formally assess AI-generated content.
+
+### Acceptable AI Content
+- Demonstrates genuine E-E-A-T despite AI assistance
+- Provides unique value (not just rephrased existing content)
+- Has clear human oversight and editing
+- Contains original analysis, insights, or perspective
+- Includes first-party data or real-world evidence
+
+### Low-Quality AI Content Markers
+- Generic phrasing with no specificity
+- Repetitive structure across multiple pages (cookie-cutter)
+- No original insights or unique data
+- Missing author attribution entirely
+- Factual inaccuracies or hallucinated statistics
+- Excessive use of filler words and hedging language
+
+---
+
+## Scoring Methodology
+
+**Total E-E-A-T Score = Experience + Expertise + Authoritativeness + Trustworthiness (out of 100)**
+
+| Range | Rating | Interpretation |
+|-------|--------|---------------|
+| 80–100 | Excellent | Strong signals across all dimensions |
+| 60–79 | Good | Solid foundation with room to improve |
+| 40–59 | Needs Work | Significant gaps in one or more areas |
+| 20–39 | Poor | Major deficiencies undermining credibility |
+| 0–19 | Critical | Lacking basic trust and expertise signals |

package/src/bmad-plus/packs/pack-seo/ref/geo-signals.md

@@ -0,0 +1,167 @@
+# GEO Signals — AI Search Optimization Guide (March 2026)
+
+> Author: Laurent Rochetta | BMAD+ SEO Engine v2.0
+
+## What is GEO?
+
+**Generative Engine Optimization (GEO)** is the discipline of optimizing content for AI-generated answers across Google AI Overviews, ChatGPT, Perplexity, Bing Copilot, and other AI-powered search experiences.
+
+## AI Search Landscape
+
+| Platform | Reach | Key Citation Sources |
+|----------|-------|---------------------|
+| Google AI Overviews | 1.5B users/month, 200+ countries, 50%+ query coverage | Top-10 ranking pages (92%) |
+| Google AI Mode | Full conversational search (May 2025), zero organic links | Citation is the only visibility mechanism |
+| ChatGPT Search | 900M weekly active users | Wikipedia (47.9%), Reddit (11.3%) |
+| Perplexity | 500M+ monthly queries | Reddit (46.7%), Wikipedia |
+| Bing Copilot | Integrated in Edge/Windows | Bing index, authoritative sites |
+
+> AI-referred sessions grew **527%** between January and May 2025 (SparkToro).
+
+---
+
+## The Brand Mention Revolution
+
+**Brand mentions correlate 3× more strongly with AI visibility than backlinks.**
+(Ahrefs December 2025 study, 75,000 brands)
+
+| Signal | Correlation with AI Citations |
+|--------|-------------------------------|
+| YouTube mentions | ~0.737 (strongest) |
+| Reddit mentions | High |
+| Wikipedia presence | High |
+| LinkedIn presence | Moderate |
+| Domain Rating (backlinks) | ~0.266 (weak) |
+
+> Only **11%** of domains are cited by both ChatGPT and Google AI Overviews for the same query — platform-specific optimization is essential.
+
+---
+
+## GEO Scoring Criteria
+
+### 1. Citability (25%)
+
+**Optimal passage length: 134–167 words** for AI citation.
+
+| Signal | Strong | Weak |
+|--------|--------|------|
+| Quotable statements with facts/stats | ✅ | Vague generic statements |
+| Self-contained answer blocks | ✅ | Requires surrounding context |
+| Direct answer in first 40–60 words | ✅ | Buried conclusions |
+| "X is..." / "X refers to..." patterns | ✅ | No definitions |
+| Unique data not found elsewhere | ✅ | Commonly available info |
+
+### 2. Structural Readability (20%)
+
+92% of AI Overview citations come from top-10 pages, but 47% come from pages below position 5 — demonstrating different selection logic.
+
+| Signal | Strong | Weak |
+|--------|--------|------|
+| Clean H1→H2→H3 hierarchy | ✅ | Wall of text |
+| Question-based headings | ✅ | Vague headings |
+| Short paragraphs (2-4 sentences) | ✅ | Long unbroken paragraphs |
+| Tables for comparative data | ✅ | No structured data display |
+| Lists for multi-item content | ✅ | Inline lists in sentences |
+
+### 3. Multi-Modal Content (15%)
+
+Content with multi-modal elements sees **156% higher selection rates**.
+
+- Text + relevant images
+- Video content (embedded or linked)
+- Infographics and charts
+- Interactive tools (calculators, assessments)
+- Structured data supporting media
+
+### 4. Authority & Brand Signals (20%)
+
+- Author byline with credentials
+- Publication and last-updated dates
+- Citations to primary sources
+- Organization credentials
+- Entity presence in Wikipedia, Wikidata
+- Active presence on Reddit, YouTube, LinkedIn
+
+### 5. Technical Accessibility (20%)
+
+**AI crawlers do NOT execute JavaScript** — SSR is critical.
+
+- Server-side rendering for core content
+- AI crawler access in robots.txt
+- llms.txt file presence
+- RSL 1.0 licensing terms
+
+---
+
+## AI Crawler Registry
+
+| Crawler | Owner | Purpose | Recommend |
+|---------|-------|---------|-----------|
+| GPTBot | OpenAI | ChatGPT web search + training | Allow |
+| OAI-SearchBot | OpenAI | OpenAI search features | Allow |
+| ChatGPT-User | OpenAI | ChatGPT browsing | Allow |
+| ClaudeBot | Anthropic | Claude web features | Allow |
+| PerplexityBot | Perplexity | Perplexity search | Allow |
+| Applebot-Extended | Apple | Apple Intelligence features | Allow |
+| Google-Extended | Google | Gemini training (NOT Search) | Optional |
+| CCBot | Common Crawl | Open dataset training | Optional/Block |
+| Bytespider | ByteDance | TikTok/Douyin AI training | Block |
+| anthropic-ai | Anthropic | Claude training | Optional |
+| cohere-ai | Cohere | Cohere models | Optional |
+
+---
+
+## llms.txt Standard
+
+File at `/llms.txt` (domain root). Provides structured guidance to AI crawlers.
+
+```
+# Site Name
+> One-line description
+
+## Main Sections
+- [Page Title](url): Description
+- [Another Page](url): Description
+
+## Key Facts
+- Important fact 1
+- Important fact 2
+```
+
+Also check for `/llms-full.txt` (expanded version).
+
+---
+
+## RSL 1.0 (Really Simple Licensing)
+
+Machine-readable AI licensing standard (December 2025).
+Backed by: Reddit, Yahoo, Medium, Quora, Cloudflare, Akamai, Creative Commons.
+
+Check for RSL implementation and appropriate licensing terms.
+
+---
+
+## Quick Wins for AI Visibility
+
+### Immediate (1 day)
+1. Add "What is [topic]?" definition in first 60 words of key pages
+2. Create 134–167 word self-contained answer blocks
+3. Add question-based H2/H3 headings
+4. Include specific statistics with source attribution
+5. Add publication and update dates
+6. Allow key AI crawlers in robots.txt (GPTBot, ClaudeBot, PerplexityBot)
+
+### Short-term (1 week)
+1. Create `/llms.txt` file
+2. Add author bios with credentials + Wikipedia/LinkedIn links
+3. Implement Person and Organization schema
+4. Add comparison tables with structured data
+5. Ensure SSR for all critical content
+
+### Strategic (1-3 months)
+1. Build Wikipedia presence for brand and key people
+2. Create original research or surveys (unique citability)
+3. Establish active YouTube channel
+4. Build Reddit community presence
+5. Implement comprehensive entity linking (sameAs across platforms)
+6. Develop unique tools, calculators, or interactive content

package/src/bmad-plus/packs/pack-seo/ref/hreflang-rules.md

@@ -0,0 +1,153 @@
+# Hreflang — Audit Rules & Best Practices (March 2026)
+
+> Author: Laurent Rochetta | BMAD+ SEO Engine v2.0
+
+## What is Hreflang?
+
+`hreflang` tells search engines which language/region version of a page to serve.
+Errors cause wrong language indexing, duplicate content penalties, and lost organic traffic.
+
+---
+
+## Implementation Methods
+
+| Method | Best For | Max Pages |
+|--------|----------|-----------|
+| `<link>` in `<head>` | Small sites (<50 pages) | ~50 |
+| HTTP header `Link:` | Non-HTML files (PDFs) | ~50 |
+| Sitemap `<xhtml:link>` | Large sites (50+) | Unlimited |
+
+> **Recommendation**: Use sitemap for sites with 50+ pages. HTTP headers for non-HTML resources.
+
+---
+
+## Validation Rules
+
+### Rule 1: Valid Language Codes
+- Use **ISO 639-1** (2-letter): `en`, `fr`, `de`, `es`, `ja`, `zh`
+- Optional **ISO 3166-1 Alpha-2** for region: `en-US`, `en-GB`, `fr-FR`, `fr-CA`, `pt-BR`
+- **Case insensitive** but convention is lowercase lang, uppercase country
+
+| ✅ Valid | ❌ Invalid | Why |
+|---------|-----------|-----|
+| `en` | `english` | Must be ISO 639-1 |
+| `fr-FR` | `fr-FRA` | Country must be 2-letter |
+| `zh-Hans` | `cn` | `cn` is not a valid language code |
+| `x-default` | `default` | Must use exact `x-default` |
+
+### Rule 2: Self-Referencing (MANDATORY)
+Every page MUST include a hreflang tag pointing to itself.
+
+```html
+<!-- On the English page (example.com/en/) -->
+<link rel="alternate" hreflang="en" href="https://example.com/en/" />
+<link rel="alternate" hreflang="fr" href="https://example.com/fr/" />
+<link rel="alternate" hreflang="x-default" href="https://example.com/en/" />
+```
+
+**Error if missing**: Google may ignore all hreflang tags on that page.
+
+### Rule 3: Return Tags (MANDATORY)
+If page A links to page B with hreflang, page B MUST link back to page A.
+
+```
+Page A (en) → hreflang="fr" → Page B (fr)
+Page B (fr) → hreflang="en" → Page A (en)   ← MANDATORY
+```
+
+**Error if missing**: Called "orphan hreflang" — Google ignores the one-way tag.
+
+### Rule 4: x-default (STRONGLY RECOMMENDED)
+Designate a fallback page for users whose language/region doesn't match any variant.
+
+```html
+<link rel="alternate" hreflang="x-default" href="https://example.com/" />
+```
+
+Common choices for x-default:
+- Language selector/redirect page
+- English version (most common)
+- Homepage of the main domain
+
+### Rule 5: Canonical + Hreflang Consistency
+- Each hreflang URL **must be the canonical version** (not a redirect, not a URL with parameters)
+- If a page has `rel="canonical"` pointing elsewhere, hreflang tags on that page are **ignored**
+- Canonical and hreflang must agree: don't have hreflang point to a URL that canonicalizes to a different URL
+
+### Rule 6: Absolute URLs Only
+```html
+<!-- ✅ Correct -->
+<link rel="alternate" hreflang="fr" href="https://example.com/fr/page" />
+
+<!-- ❌ Wrong -->
+<link rel="alternate" hreflang="fr" href="/fr/page" />
+```
+
+### Rule 7: No Hreflang on Non-200 Pages
+- Don't include hreflang tags on pages that return 3xx, 4xx, or 5xx
+- Don't point hreflang to pages that redirect
+
+---
+
+## Common Error Patterns
+
+### Error 1: Missing Return Tags
+**Symptom**: hreflang is configured on the main language but not on alternate versions.
+**Fix**: Add reciprocal hreflang tags on ALL language variants.
+
+### Error 2: Wrong Canonical + Hreflang
+**Symptom**: Page A hreflang → Page B, but Page B canonical → Page C.
+**Fix**: Align canonical and hreflang targets.
+
+### Error 3: Missing Self-Reference
+**Symptom**: Page lists other language versions but not itself.
+**Fix**: Add `hreflang` tag with the current page's own language/URL.
+
+### Error 4: Inconsistent URLs
+**Symptom**: hreflang uses `http://` but site is on `https://`, or trailing slash mismatch.
+**Fix**: Use exact canonical URL (protocol, www/non-www, trailing slash).
+
+### Error 5: Language vs Region Confusion
+**Symptom**: Using `hreflang="fr"` for France and `hreflang="fr"` for Canada.
+**Fix**: Use `hreflang="fr-FR"` and `hreflang="fr-CA"` to differentiate.
+
+### Error 6: Missing x-default
+**Symptom**: Users in unsupported regions see random language version.
+**Fix**: Add `x-default` pointing to language selector or English version.
+
+---
+
+## Sitemap Implementation (Recommended for 50+ pages)
+
+```xml
+<?xml version="1.0" encoding="UTF-8"?>
+<urlset xmlns="http://www.sitemaps.org/schemas/sitemap/0.9"
+        xmlns:xhtml="http://www.w3.org/1999/xhtml">
+  <url>
+    <loc>https://example.com/en/page</loc>
+    <xhtml:link rel="alternate" hreflang="en" href="https://example.com/en/page"/>
+    <xhtml:link rel="alternate" hreflang="fr" href="https://example.com/fr/page"/>
+    <xhtml:link rel="alternate" hreflang="de" href="https://example.com/de/page"/>
+    <xhtml:link rel="alternate" hreflang="x-default" href="https://example.com/en/page"/>
+  </url>
+</urlset>
+```
+
+---
+
+## Audit Checklist
+
+| # | Check | Priority |
+|---|-------|----------|
+| 1 | All hreflang language codes are valid ISO 639-1 | 🔴 Critical |
+| 2 | All hreflang country codes are valid ISO 3166-1 | 🔴 Critical |
+| 3 | Every page has self-referencing hreflang | 🔴 Critical |
+| 4 | All hreflang tags have return tags | 🔴 Critical |
+| 5 | All hreflang URLs are absolute | 🔴 Critical |
+| 6 | x-default is specified | 🟠 High |
+| 7 | Hreflang URLs match canonical URLs | 🟠 High |
+| 8 | No hreflang on non-200 pages | 🟠 High |
+| 9 | No hreflang pointing to redirecting URLs | 🟠 High |
+| 10 | Consistent protocol (https) and www/non-www | 🟡 Medium |
+| 11 | Language/region differentiation correct | 🟡 Medium |
+| 12 | Sitemap implementation for 50+ pages | 🟡 Medium |

package/src/bmad-plus/packs/pack-seo/ref/quality-gates.md

@@ -0,0 +1,133 @@
+# Quality Gates — Content Thresholds (March 2026)
+
+> Author: Laurent Rochetta | BMAD+ SEO Engine v2.0
+
+## Word Count Minimums by Page Type
+
+| Page Type | Minimum | Notes |
+|-----------|---------|-------|
+| Homepage | 500 | Brand clarity + key offerings |
+| Service page | 800 | Comprehensive service description |
+| Blog / article | 1,500 | Deep topical coverage |
+| Product page | 300+ (400+ complex) | Depends on product complexity |
+| Category page | 200 | Plus product listings |
+| Location page | 500–600 | Must be 60%+ unique per location |
+| Comparison page | 1,200 | Feature matrix + analysis |
+| Landing page | 400 | Conversion-focused |
+| About page | 500 | Company story + credentials |
+| FAQ page | 1,000 | Comprehensive Q&A coverage |
+
+> **Word count is NOT a ranking factor.** These are topical coverage floors. A thorough 500-word page beats a padded 2,000-word one.
+
+---
+
+## Location Page Quality Gates
+
+| Threshold | Action |
+|-----------|--------|
+| 1–29 location pages | ✅ Normal — ensure 60%+ unique content per page |
+| 30+ location pages | ⚠️ WARNING — enforce 60%+ unique content, audit for thin/duplicate |
+| 50+ location pages | 🛑 HARD STOP — require user justification before proceeding |
+
+### Unique Content Requirements per Location
+- Unique local testimonials or case studies
+- Location-specific service details
+- Local area information and context
+- Original photos (not stock)
+- Unique meta title and description
+
+---
+
+## Programmatic Page Quality Gates
+
+| Threshold | Action |
+|-----------|--------|
+| 1–99 programmatic pages | ✅ Normal — verify template quality |
+| 100+ pages | ⚠️ WARNING — audit for thin content and cannibalization |
+| 500+ pages | 🛑 HARD STOP — full audit required before generating more |
+
+### Programmatic Content Requirements
+- Each page must provide unique value (not just data swap)
+- Internal linking automation between related pages
+- Canonical strategy to prevent index bloat
+- Minimum 200 words of unique contextual content per page
+- Quality varies by template — audit samples before scale
+
+---
+
+## Title Tag Guidelines
+
+| Criterion | Requirement |
+|-----------|-------------|
+| Length | 50–60 characters (display limit ~600px) |
+| Primary keyword | Include near the beginning |
+| Brand name | Append with " | Brand" or " — Brand" |
+| Uniqueness | Every page must have a unique title |
+| Accuracy | Must reflect actual page content |
+
+---
+
+## Meta Description Guidelines
+
+| Criterion | Requirement |
+|-----------|-------------|
+| Length | 150–160 characters |
+| Call to action | Include when appropriate |
+| Primary keyword | Include naturally |
+| Uniqueness | Every page must have a unique description |
+| Accuracy | Must match page content (Google may rewrite if not) |
+
+---
+
+## Heading Hierarchy Rules
+
+| Rule | Requirement |
+|------|-------------|
+| H1 count | Exactly 1 per page |
+| H1 content | Should contain primary keyword, be descriptive |
+| Hierarchy | H1 → H2 → H3 (no skipping levels) |
+| Heading frequency | One heading every 200–300 words |
+| Heading style | Descriptive, not generic ("Our Services" → "Professional SEO Audit Services") |
+
+---
+
+## Image Optimization Standards
+
+| Criterion | Standard | Priority |
+|-----------|----------|----------|
+| Alt text | Descriptive, meaningful, includes keywords naturally | High |
+| Dimensions | Width and height attributes specified | High (CLS) |
+| Format | WebP or AVIF preferred, JPEG/PNG acceptable | Medium |
+| File size | <200KB for standard, <500KB for hero images | Medium |
+| Lazy loading | `loading="lazy"` for below-fold images | Medium |
+| Preload | `<link rel="preload">` for hero/LCP images | High |
+| Srcset | Responsive sizes for different viewports | Medium |
+
+---
+
+## Internal Linking Standards
+
+| Metric | Target |
+|--------|--------|
+| Links per 1,000 words | 3–5 relevant internal links |
+| Anchor text | Descriptive (not "click here") |
+| Orphan pages | 0 (every page linked from at least one other) |
+| Click depth | Important pages within 3 clicks of homepage |
+| Broken links | 0 internal 404s |
+
+---
+
+## Thin Content Detection
+
+A page is flagged as "thin content" when:
+1. Word count is below the minimum for its page type
+2. Content is predominantly duplicate or near-duplicate of another page
+3. Content is auto-generated without unique value-add
+4. Page provides no actionable information
+
+### Doorway Page Prevention
+Never recommend creating pages that:
+- Exist solely to funnel users to another page
+- Have nearly identical content with only location/keyword swaps
+- Provide no standalone value
+- Target many similar keyword variations with minimal differences