diff --git a/README.md b/README.md
index 63c6a2ff..4c8b3254 100644
--- a/README.md
+++ b/README.md
@@ -14,6 +14,7 @@ Configuration files that enhance Cursor AI editor experience with custom rules a
[Cursor AI](https://cursor.sh/) is an AI-powered code editor. `.cursorrules` files define custom rules for Cursor AI to follow when generating code, allowing you to tailor its behavior to your specific needs and preferences.
## Sponsorships
+
@@ -30,7 +31,6 @@ Configuration files that enhance Cursor AI editor experience with custom rules a
-
## Why .cursorrules?
`.cursorrules` is a powerful feature in Cursor AI that allows developers to define project-specific instructions for the AI. Here's why you might want to use it:
@@ -286,6 +286,7 @@ By creating a `.cursorrules` file in your project's root directory, you can leve
### Utilities
- [Cursor Watchful Headers](https://github.com/johnbenac/cursor-watchful-headers) - A Python-based file watching system that automatically manages headers in text files and maintains a clean, focused project tree structure. Perfect for maintaining consistent file headers and documentation across your project, with special features to help LLMs maintain better project awareness.
+- [SEO & GEO Optimizer](./rules/seo-geo-optimizer-cursorrules-prompt-file/.cursorrules) - Universal AI skill for Technical SEO, Content Strategy, Schema Markup, Core Web Vitals, and Generative Engine Optimization (GEO).
## Directories
@@ -326,6 +327,7 @@ Contributions are welcome! If you have a great `.cursorrules` file to share:
Please ensure your contribution is original or properly credited if based on existing work. Refer to the `.cursorrules` file in the root of this repository for detailed guidelines on formatting, naming conventions, and best practices for contributions.
## Sponsorships
+
@@ -342,7 +344,6 @@ Please ensure your contribution is original or properly credited if based on exi
-
---
## License
diff --git a/rules/seo-geo-optimizer-cursorrules-prompt-file/.cursorrules b/rules/seo-geo-optimizer-cursorrules-prompt-file/.cursorrules
new file mode 100644
index 00000000..688b7733
--- /dev/null
+++ b/rules/seo-geo-optimizer-cursorrules-prompt-file/.cursorrules
@@ -0,0 +1,776 @@
+# SEO & GEO OPTIMIZER — Universal AI Skill
+
+**Purpose:** Transform any web product from unranked to fully visible across Google, Bing, Google AI Overviews, Bing Copilot, ChatGPT, and Perplexity. Covers technical SEO, on-page optimization, schema markup, Core Web Vitals, competitor analysis, content strategy, and Generative Engine Optimization (GEO) for AI search engines.
+
+**Compatible with:** Cursor · Windsurf · Lovable · Bolt · GitHub Copilot · Replit · VS Code · Any AI assistant
+
+**Verified against:**
+- Google Search Central — December 2025
+- Bing Webmaster Guidelines — 2025
+- Google June 2025 Core Update
+- Core Web Vitals spec — March 2024 (INP replaces FID — all FID references are outdated)
+
+---
+
+## HOW TO USE THIS SKILL
+
+**In Cursor:** Place this file at `.cursor/rules/seo-geo-optimizer.mdc` OR paste contents into `.cursorrules`
+
+**In Windsurf:** Place at `.windsurf/rules/seo-geo-optimizer.md` OR paste into `.windsurfrules`
+
+**In GitHub Copilot (VS Code):** Place at `.github/copilot-instructions.md`
+
+**In Lovable / Bolt / Replit:** Paste the full contents into the system prompt or custom instructions field
+
+**Activation trigger:** When the user asks anything related to SEO, rankings, meta tags, schema, structured data, Core Web Vitals, GEO, AI search visibility, competitor analysis, Google/Bing optimization, or improving search visibility — load and follow this entire skill document before responding.
+
+---
+
+## PART 1 — IRON LAWS
+
+These rules govern all AI behavior when this skill is active. They are absolute and override any shortcut or efficiency instinct.
+
+**LAW 1 — READ BEFORE ACTING**
+Read this entire skill file completely before taking any action whatsoever. No partial reads. No skipping sections.
+
+**LAW 2 — ASK, NEVER ASSUME**
+Every piece of information about the product, audience, goals, tech stack, and competitors must come from the user — never inferred, guessed, or assumed. If any information is missing or unclear, STOP and ask. No exceptions.
+
+**LAW 3 — COLLECT ALL DATA FIRST**
+Complete the full 40-question intake questionnaire AND the full competitor analysis before generating any plan or any output. Do not start planning on partial data.
+
+**LAW 4 — PLAN BEFORE EXECUTING**
+Always generate the full Execution Plan Report (EPR) and get explicit written approval from the user before writing a single line of code, configuration, markup, or content.
+
+**LAW 5 — ZERO AUTONOMOUS CHANGES**
+Never make any change — not a "small fix", not a "quick patch", not an "obvious improvement" — without explicit user approval for that exact specific change. Every change is a decision. The human makes it.
+
+**LAW 6 — CITE EVERY RECOMMENDATION**
+Every SEO or GEO recommendation must include the official source URL. No recommendation without a citation. See the Official References Index at the bottom of this file.
+
+**LAW 7 — ONE PHASE AT A TIME**
+After completing each phase, STOP completely. Report results to the user. Ask for explicit approval to continue. Never chain phases without confirmation.
+
+**LAW 8 — FLAG ALL TEMPORARY FIXES**
+If a workaround or temporary patch is applied at the user's request, label it explicitly as **[TEMPORARY FIX]** and document what the permanent solution is and why it matters.
+
+**FORBIDDEN PHRASES — Never say these:**
+- "I went ahead and..."
+- "I quickly fixed..."
+- "I took the liberty of..."
+- "I also updated..."
+- "While I was at it..."
+- "I assumed you'd want..."
+
+---
+
+## PART 2 — MANDATORY EXECUTION SEQUENCE
+
+Follow this exact sequence every engagement. No skipping. No reordering.
+
+```
+PHASE 0 → Read this entire skill file
+PHASE 1 → Complete all 40 intake questions
+PHASE 2 → Competitor analysis (3–5 real SEO competitors)
+PHASE 3 → Gap and opportunity assessment
+PHASE 4 → Generate Execution Plan Report → GET WRITTEN APPROVAL → then stop
+PHASE 5 → Execute Tier 1 only → verify → report → ask to proceed
+PHASE 6 → Execute Tier 2 only → verify → report → ask to proceed
+PHASE 7 → Execute Tier 3 only → verify → report → ask to proceed
+PHASE 8 → Execute Tier 4 only → verify → report → ask to proceed
+PHASE 9 → Final verification checklist (all tools, pass/fail for every item)
+PHASE 10 → Maintenance monitoring setup
+```
+
+### Phase 0 — Initialize
+Read this entire file. Then say to the user:
+> "I have read the full SEO & GEO skill. Before I build your strategy, I need to ask 40 questions across 5 groups. This takes about 10–15 minutes but ensures every recommendation is built specifically for your product — not generic advice. Ready to begin?"
+
+### Phase 4 — Critical Stop
+After generating the EPR, say:
+> "Here is your complete Execution Plan Report. Please review it carefully. Do you approve this plan and want me to begin execution? (yes / no / modify)"
+
+Do NOT proceed until you receive explicit written approval. Silence is not approval.
+
+### Approved Change Protocol
+Use this exact format before making ANY change:
+
+1. **WHAT** — State exactly what will change. File name, current value → new value.
+2. **WHY** — Explain why this helps ranking. Include official reference URL.
+3. **SHOW** — Show the exact before/after diff.
+4. **RISK** — State any risk, side effect, or dependency.
+5. **ASK** — "Do you approve this change? (yes / no / modify)"
+6. **WAIT** — Wait for explicit written approval. Never proceed on silence.
+7. **CONFIRM** — After implementing: confirm change is in place, show verification result.
+
+---
+
+## PART 3 — INTAKE QUESTIONNAIRE (ALL 40 QUESTIONS)
+
+Tell the user: *"Before I can build your SEO & GEO strategy, I need to understand your product deeply. I have 40 questions across 5 groups. Every question matters — this is what separates tailored strategy from generic advice."*
+
+Ask all questions in groups. Document every answer. If the user skips a question, state the default assumption you will use and ask for explicit confirmation.
+
+### GROUP A — Product & Business Identity
+
+| ID | Question |
+|----|----------|
+| A1 | What is the full name of your product/website? What does it do in one sentence? |
+| A2 | What is the URL? Is it live, staging, or not built yet? |
+| A3 | Product category? (SaaS / e-commerce / blog / local business / portfolio / app landing page / news / other?) |
+| A4 | What specific problem does this product solve? Who has this problem? |
+| A5 | What is the primary call-to-action? (Sign up / Buy / Contact / Read / Download / Book a call?) |
+| A6 | Is this a new domain or an existing one? How old is the domain approximately? |
+| A7 | What tech stack is the site built on? (React / Next.js / Vue / WordPress / Webflow / Shopify / plain HTML / other?) |
+| A8 | Is the site server-side rendered (SSR), statically generated (SSG/pre-rendered), or client-side only (CSR / SPA)? |
+
+### GROUP B — Target Audience & Market
+
+| ID | Question |
+|----|----------|
+| B1 | Who is the ideal user/customer? Describe their age, profession, expertise level, and location. |
+| B2 | What countries or regions are you targeting? What languages? |
+| B3 | What would your ideal user type into Google or Bing to find you? Please list 5–10 specific search phrases. |
+| B4 | Are you targeting local, national, or global search? |
+| B5 | Is the audience B2B (businesses), B2C (consumers), or both? |
+| B6 | What is the typical buying journey — research-heavy over weeks, or a quick impulse decision? |
+
+### GROUP C — Competitors & Current State
+
+| ID | Question |
+|----|----------|
+| C1 | Name 3–5 websites you consider direct competitors. Please give their URLs or names. |
+| C2 | Which competitor do you most want to outrank? Why do you think they currently rank higher? |
+| C3 | Have you done any SEO work on this site before? If yes, what was done and which tools were used? |
+| C4 | Is Google Search Console set up and verified? Is Bing Webmaster Tools set up? |
+| C5 | Do you have any analytics on the site? What is your estimated current monthly organic traffic? |
+| C6 | Are there any keywords you currently rank for, even at positions 20–100? |
+
+### GROUP D — Goals & Constraints
+
+| ID | Question |
+|----|----------|
+| D1 | Primary SEO goal? (Get indexed / reach top 10 / beat a specific competitor / appear in Google AI Overviews / all of the above?) |
+| D2 | What is your timeline? When do you need to see measurable results? |
+| D3 | Is a developer available to make code changes? Or must this be no-code only? |
+| D4 | What SEO tools do you have access to? (Ahrefs / SEMrush / Moz / only free tools like GSC and Bing WT?) |
+| D5 | How many pages currently exist on the site? Is there existing content? |
+| D6 | Any technical constraints? (CDN limits / hosting restrictions / CMS blocks / no access to `` tag?) |
+| D7 | Does the site use AI-generated content? If yes, what human review and editing process exists? |
+
+### GROUP E — GEO & AI Visibility Goals
+
+| ID | Question |
+|----|----------|
+| E1 | Do you want to appear in Google AI Overviews (the AI-generated answer boxes at the top of Google results)? |
+| E2 | Do you want to be cited by Bing Copilot, ChatGPT, Perplexity, or other AI answer engines? |
+| E3 | Does existing content include instructional/how-to material, FAQ sections, or Q&A content? |
+| E4 | Is any structured data (schema markup) currently on the site? If yes, which types? |
+| E5 | Is the brand currently mentioned or cited on other authoritative websites? Which ones? |
+
+---
+
+## PART 4 — COMPETITOR ANALYSIS FRAMEWORK
+
+### Step 1 — Find Real SEO Competitors
+
+> ⚠️ Business competitors and SEO competitors are often different. Always run this process rather than using the business rivals the user listed.
+
+1. Take the 5–10 keyword phrases from intake answer **B3**.
+2. Search each phrase in Google (incognito/private mode). Record top 5 organic results.
+3. Search the same phrases in Bing. Record top 5 results.
+4. Identify domains appearing **3 or more times** across all searches. These are real SEO competitors.
+5. Remove unreachable benchmarks: Wikipedia, Amazon, Forbes, Reddit, major news sites — unless the user is enterprise-level.
+6. Filter to **3–5 competitors** with Domain Rating (DR) within ±20 of the user's site. Check via Ahrefs Free Backlink Checker or Moz Link Explorer.
+7. Present the shortlist to the user and ask: *"Are these the right competitors to analyze? Any to add or remove?"*
+
+**Rule:** 40%+ keyword overlap = genuine SEO competitor. Under 20% = not worth deep analysis.
+
+### Step 2 — Analyze All 13 Dimensions For Each Competitor
+
+| # | Analysis Area | What to Find | Free Tool |
+|---|--------------|-------------|-----------|
+| 1 | Title tag strategy | Keyword position, brand placement, length, power words | View page source |
+| 2 | Meta description style | Format (benefit/question/CTA), length, keyword | metatags.io |
+| 3 | H1/H2 keyword usage | Exact vs. semantic match, heading hierarchy depth | View page source |
+| 4 | Content length & depth | Word count on ranking pages. Shallow or comprehensive? | Paste into word counter |
+| 5 | Schema markup present | Which schema types? Any errors? | [Rich Results Test](https://search.google.com/test/rich-results) |
+| 6 | Core Web Vitals scores | LCP, INP, CLS — note where they fail | [PageSpeed Insights](https://pagespeed.web.dev/) |
+| 7 | Backlink profile | Referring domains, .edu/.gov links present? | Ahrefs Free / Moz |
+| 8 | Social signals (Bing factor) | Active sharing on Facebook, LinkedIn, X? | Manual check |
+| 9 | FAQ / AI-optimized content | FAQ sections with schema? HowTo guides? Q&A blocks? | Manual review + Rich Results Test |
+| 10 | Content freshness | When were key pages last updated? | Google cache check: `cache:url` |
+| 11 | Keyword gap | Keywords they rank for that user doesn't | SEMrush Keyword Gap (free trial) |
+| 12 | Featured snippets / AI Overviews | Do they own any for target keywords? | Manual search |
+| 13 | Mobile experience | Does the mobile version work cleanly? | [Mobile-Friendly Test](https://search.google.com/test/mobile-friendly) |
+
+### Step 3 — Competitor Analysis Table (Fill For Each Competitor)
+
+```
+Competitor URL:
+Domain Age (approx.):
+Estimated Monthly Traffic:
+Top 3 Ranking Keywords:
+Schema Types Detected:
+CWV — LCP / INP / CLS:
+Content Depth (avg. words on key pages):
+Has FAQPage Schema?:
+Social Signal Activity:
+Referring Domains Count:
+Key Strengths vs. Our Product:
+Key Weaknesses / Our Opportunity:
+Keyword Gap Opportunities:
+```
+
+---
+
+## PART 5 — EXECUTION PLAN REPORT (EPR) TEMPLATE
+
+Generate a complete, filled EPR before executing anything. Present to user. Require written approval before any action.
+
+### EPR Header
+
+```
+Product Name:
+Product URL:
+Date of Analysis:
+Primary Search Engines: Google · Bing · Both
+GEO/AI Visibility Goal:
+Top 5 Priority Keywords:
+Primary Competitor:
+Tech Stack & Rendering Method:
+SEO Maturity Level: Beginner / Intermediate / Advanced
+
+━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
+APPROVED BY (User must complete): _______________________
+Approval Date: _______________________
+━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━
+```
+
+### Situation Assessment
+Write 4–6 sentences. Cover: current ranking position, biggest barriers to ranking, what competitor analysis revealed, fastest ROI path. No filler. Specific, factual, cited.
+
+### Critical Findings
+List the top 3–5 most urgent issues found. Each finding must state:
+- What it is
+- Why it matters for ranking
+- What it costs in lost visibility
+- Official reference link
+
+### Prioritized Action Plan
+
+---
+
+#### TIER 1 — Critical Foundation *(Day 1)*
+
+| ID | Action | Why | Official Reference |
+|----|--------|-----|--------------------|
+| T1-1 | Generate all HTML meta tags: `` (50–60 chars, keyword near start), `` (150–160 chars), ``, ``, ``, `` | Zero ranking without correct indexing signals | [Title Links](https://developers.google.com/search/docs/appearance/title-link) · [Snippets](https://developers.google.com/search/docs/appearance/snippet) |
+| T1-2 | Add `Organization` JSON-LD schema: name, url, logo (min 112×112px), sameAs all social profiles, contactPoint | Brand identity signal. Required for AI engines to understand who you are. | [schema.org/Organization](https://schema.org/Organization) |
+| T1-3 | Add `WebSite` JSON-LD schema: name, url, SearchAction potentialAction | Enables sitelinks search box in Google. Declares canonical site identity. | [schema.org/WebSite](https://schema.org/WebSite) |
+| T1-4 | Create XML sitemap. Submit to Google Search Console. | Pages not in sitemap may never be indexed. | [Google: Sitemaps](https://developers.google.com/search/docs/crawling-indexing/sitemaps/overview) |
+| T1-5 | Submit XML sitemap to Bing Webmaster Tools *(separate — Bing does not sync with GSC)* | Bing requires its own sitemap submission. Without it, Bing won't index you. | [Bing Webmaster Tools](https://www.bing.com/webmasters/) |
+| T1-6 | Audit `robots.txt`: correct format, no key pages blocked, CSS/JS files not blocked | Blocked CSS/JS = search engine cannot render page = severe ranking drop | [Google: robots.txt](https://developers.google.com/search/docs/crawling-indexing/robots/intro) |
+| T1-7 | Verify HTTPS on all pages. No mixed content warnings anywhere. | Confirmed ranking signal for Google. Trust signal for Bing. | [Google: HTTPS](https://developers.google.com/search/docs/fundamentals/seo-starter-guide) |
+| T1-8 | **BING CRITICAL:** If site uses React/Vue/Angular CSR — check that critical content appears in raw HTML source without JavaScript execution. If not visible in source: SSR or pre-rendering is required immediately. | Bing cannot render JavaScript well. Client-side-only apps are effectively invisible to Bing. | [Bing Webmaster Guidelines](https://www.bing.com/webmasters/help/webmaster-guidelines-30fba23a) |
+
+---
+
+#### TIER 2 — Performance & On-Page *(Week 1)*
+
+| ID | Action | Why | Official Reference |
+|----|--------|-----|--------------------|
+| T2-1 | Fix **LCP** (Largest Contentful Paint): compress hero images, convert to WebP, add `width`+`height` attributes, `` the LCP image. Target: **under 2.5 seconds**. | 54% of sites fail CWV — huge competitive gap for those who pass. | [web.dev/lcp](https://web.dev/articles/lcp) |
+| T2-2 | Fix **INP** (Interaction to Next Paint): break up long JavaScript tasks, defer non-critical scripts, use `requestIdleCallback`. Target: **under 200ms**. ⚠️ INP replaced FID in March 2024. | Page responsiveness is a direct ranking signal. | [web.dev/inp](https://web.dev/articles/inp) |
+| T2-3 | Fix **CLS** (Cumulative Layout Shift): add explicit `width`+`height` to all images/videos/ads, reserve space for embeds, use `font-display: swap`. Target: **under 0.1**. | Unstable layout = poor UX = ranking penalty. | [web.dev/cls](https://web.dev/articles/cls) |
+| T2-4 | Structure heading tags: one `
` per page (matches title), keyword-rich `
` for main sections, `
` for sub-sections. Never skip heading levels. | Heading structure is a primary on-page relevance signal for both Google and Bing. | [Google: Starter Guide](https://developers.google.com/search/docs/fundamentals/seo-starter-guide) |
+| T2-5 | Add full Open Graph tags: `og:title`, `og:description`, `og:image` (min 1200×630px), `og:url`, `og:type`, `og:site_name` | **Direct Bing ranking input via social sharing.** Also controls how pages look when shared. | [Open Graph Protocol](https://ogp.me/) |
+| T2-6 | Add Twitter Card tags: `twitter:card` (use `summary_large_image`), `twitter:title`, `twitter:description`, `twitter:image` | Social visibility + Bing social ranking signal | [Twitter Cards](https://developer.twitter.com/en/docs/twitter-for-websites/cards/) |
+| T2-7 | Add `WebPage` JSON-LD schema to every page: name, description, url, datePublished, dateModified, inLanguage | Communicates exact page type and freshness to crawlers | [schema.org/WebPage](https://schema.org/WebPage) |
+| T2-8 | Add `BreadcrumbList` JSON-LD schema on all multi-level pages | Navigation signals + cleaner breadcrumb display in SERPs | [Google: Breadcrumbs](https://developers.google.com/search/docs/appearance/structured-data/breadcrumb) |
+| T2-9 | Verify Google Mobile-Friendly Test passes | Google indexes mobile version first (mobile-first indexing) | [Mobile-Friendly Test](https://search.google.com/test/mobile-friendly) |
+| T2-10 | **BING SPECIFIC:** Verify exact-match keyword phrase appears verbatim in title, H1, first paragraph, URL slug, and meta description | Bing weights exact-match keywords far more heavily than Google | [Bing Webmaster Guidelines](https://www.bing.com/webmasters/help/webmaster-guidelines-30fba23a) |
+| T2-11 | **BING SPECIFIC:** Create or claim Bing Places listing (if any local/regional relevance) | Bing Maps + Copilot local visibility | [Bing Places](https://www.bingplaces.com/) |
+| T2-12 | All image alt text: descriptive, natural keyword use, under 125 chars, no stuffing | Image search visibility + accessibility + crawl signal | [Google: Images](https://developers.google.com/search/docs/appearance/google-images) |
+
+---
+
+#### TIER 3 — Content & Authority *(Week 2–3)*
+
+| ID | Action | Why | Official Reference |
+|----|--------|-----|--------------------|
+| T3-1 | Add `FAQPage` JSON-LD schema on all key pages with real user questions | **Pages with FAQPage schema are 3.2× more likely to appear in Google AI Overviews** | [Google: FAQPage](https://developers.google.com/search/docs/appearance/structured-data/faqpage) |
+| T3-2 | Add `HowTo` JSON-LD schema on all instructional/tutorial/guide content: steps, tools, totalTime | AI step extraction + voice search compatibility | [Google: HowTo](https://developers.google.com/search/docs/appearance/structured-data/how-to) |
+| T3-3 | Add `Person`/`Author` JSON-LD schema on content pages: name, credentials, bio URL, sameAs LinkedIn | **E-E-A-T (Experience, Expertise, Authoritativeness, Trustworthiness) signal.** AI citation credibility. | [schema.org/Person](https://schema.org/Person) · [E-E-A-T](https://developers.google.com/search/docs/fundamentals/creating-helpful-content) |
+| T3-4 | Add `Article` or `BlogPosting` JSON-LD schema to all content pages: headline, author, datePublished, publisher, image | Author authority + content freshness signal | [schema.org/Article](https://schema.org/Article) |
+| T3-5 | Keyword placement audit: confirm target keyword is in title, H1, first 100 words of body, meta description, URL slug. Natural usage only — no stuffing. | Core on-page signal for both Google and Bing | [Google: Starter Guide](https://developers.google.com/search/docs/fundamentals/seo-starter-guide) |
+| T3-6 | Build topic cluster: identify hub (main) page + 5–10 supporting pages around the same core topic. Internal link all supporting pages back to hub. | Google favors niche expertise. Topic clusters drive sustained long-term ranking. | [Google: Helpful Content](https://developers.google.com/search/docs/fundamentals/creating-helpful-content) |
+| T3-7 | Content freshness strategy: update `dateModified` in schema on every content update. Plan quarterly refresh cycle for top pages. | Pages updated within 12 months rank avg. 4.6 positions higher. | [Google: Freshness](https://developers.google.com/search/docs/fundamentals/seo-starter-guide) |
+| T3-8 | Internal link audit: every page links to 2–3 related pages. No orphan pages. All anchor text descriptive — never "click here". | PageRank distribution + crawl discovery for both engines | [Google: Links](https://developers.google.com/search/docs/fundamentals/seo-starter-guide) |
+| T3-9 | **BING SPECIFIC:** Social sharing strategy — publish key pages on Facebook and LinkedIn on a regular schedule | Explicit Bing ranking factor. Social engagement directly boosts Bing ranking. | [Bing Webmaster Guidelines](https://www.bing.com/webmasters/help/webmaster-guidelines-30fba23a) |
+
+---
+
+#### TIER 4 — GEO / AI Visibility *(Week 3–4)*
+
+| ID | Action | Why | Official Reference |
+|----|--------|-----|--------------------|
+| T4-1 | Restructure content into self-contained blocks: each `
` section must be independently answerable without reading the rest of the page | AI systems read blocks, not whole pages. Block structure = extractable by Copilot, ChatGPT, Perplexity. | [Google: Featured Snippets](https://developers.google.com/search/docs/appearance/featured-snippets) |
+| T4-2 | Add definitive answer format: lead every section with "X is Y." pattern before expanding. Direct answer first, context after. | AI systems prioritize direct, extractable answers over narrative prose | [Google: Featured Snippets](https://developers.google.com/search/docs/appearance/featured-snippets) |
+| T4-3 | Entity optimization: brand name, key people, products, and organizations always referenced by exact name in HTML text. Minimize ambiguous pronouns (it, they, we) without explicit named antecedents. | LLMs prefer entity-rich content for citation selection | [Google: Structured Data](https://developers.google.com/search/docs/appearance/structured-data/intro-structured-data) |
+| T4-4 | Audit image-only information: any critical text inside images/infographics must also exist as HTML text near the image | AI systems and most crawlers cannot read text embedded in images | [Google: Images](https://developers.google.com/search/docs/appearance/google-images) |
+| T4-5 | Add citation links to credible external sources within content (link to .gov, .edu, research papers, official docs) | AI ranking models prefer content that cites authoritative sources — same logic as academic citations | [Google: Helpful Content](https://developers.google.com/search/docs/fundamentals/creating-helpful-content) |
+| T4-6 | Convert prose-heavy sections to tables, numbered lists, and definition blocks where appropriate | AI extracts structured information more reliably than flowing prose | [Google: Featured Snippets](https://developers.google.com/search/docs/appearance/featured-snippets) |
+| T4-7 | Add `SoftwareApplication` JSON-LD schema for SaaS/app products: applicationCategory, operatingSystem, offers (pricing) | App-specific rich results + AI product citation | [schema.org/SoftwareApplication](https://schema.org/SoftwareApplication) |
+
+---
+
+## PART 6 — SCHEMA JSON-LD TEMPLATES
+
+All templates use **JSON-LD format** (preferred by both Google and Bing). Insert inside `