SEO/GEO/AEO Optimizer

Analysis and optimization toolkit for content visibility across search engines (SEO), AI platforms (GEO - Generative Engine Optimization), answer engines (AEO - Answer Engine Optimization), voice assistants, and social media.

Author: Boris Djordjevic, 199 Biotechnologies
Status: Phase 2 Complete (Content Implementation)
License: MIT
Repository: github.com/199-biotechnologies/claude-skill-seo-geo-optimizer

---

Purpose

This skill audits and optimizes content for discoverability across:

• Search engines: Google, Bing (traditional SEO)
• AI platforms: ChatGPT, Perplexity, Claude, Gemini, Grokipedia (GEO - Generative Engine Optimization)
• Answer engines: Google AI Overviews, Bing Copilot, featured snippets (AEO - Answer Engine Optimization)
• Voice assistants: Google Assistant, Siri, Alexa
• Social media: Facebook, Twitter, LinkedIn, WhatsApp, Instagram

Supports HTML, Markdown, and React/JSX files.

---

Key Statistics (2025)

AI-referred traffic grew 527% between January and May 2025. Content optimized for AI citation shows 33-40% higher visibility. Key findings:

• 40.58% of AI citations come from top 10 SERP results
• Content with author credentials: +40% citation probability
• Featured snippets: 40.7% of voice search answers
• Statistics addition: +41% citation improvement
• Freshness (<30 days): 3.2x more citations

---

Installation

Clone to Claude skills directory:

cd ~/.claude/skills/
git clone https://github.com/199-biotechnologies/claude-skill-seo-geo-optimizer.git

Requirements: Python 3.7+, no external dependencies (stdlib only).

Verify installation:

python ~/.claude/skills/seo-geo-optimizer/scripts/analyze_content.py --help

---

Usage

Phase 1: Analysis

Audit existing content, generate recommendations:

# Comprehensive audit with all reports
python scripts/audit_report.py ~/project/page.html --format all

# Analyze specific aspects
python scripts/analyze_content.py ~/project/page.html
python scripts/metadata_validator.py ~/project/page.html
python scripts/keyword_analyzer.py ~/project/page.html
python scripts/entity_extractor.py ~/project/page.html

Phase 2: Implementation

Apply optimizations automatically:

# Full optimization pipeline for specific platform
python scripts/auto_implementer.py page.html perplexity

# Individual optimization steps
python scripts/content_optimizer.py page.html
python scripts/platform_optimizer.py page.html chatgpt
python scripts/voice_optimizer.py page.html
python scripts/freshness_monitor.py page.html
python scripts/citation_enhancer.py page.html

Keyword Clustering

Group related keywords into semantic clusters for topical authority:

# Analyze with keyword clusters (default)
python scripts/keyword_analyzer.py ~/project/page.html

# Fast mode without clustering
python scripts/keyword_analyzer.py ~/project/page.html --no-clusters

Output includes:

• Semantic clusters grouped by TF-IDF similarity
• Topic identification for each cluster
• Recommendations for pillar content strategy

IndexNow Instant Indexing

Submit URLs directly to search engines for immediate indexing (critical for GEO/AEO):

# Generate IndexNow key
python scripts/indexnow_submit.py --generate-key --output ./public

# Submit single URL
python scripts/indexnow_submit.py https://example.com/new-page --key YOUR_KEY

# Batch submit from file
python scripts/indexnow_submit.py --batch urls.txt --key YOUR_KEY

Benefits for GEO:

• Bing index feeds AI platforms (ChatGPT, Perplexity, Claude)
• Fresh content indexed in minutes vs weeks
• 3.2x citation boost for content <30 days old

Generate Schema Markup

# FAQ schema (highest AI citation probability)
python scripts/schema_generator.py faq \
  --question "What is longevity medicine?" \
  --answer "Longevity medicine optimizes biomarkers like LDL <70 mg/dL to reduce cardiovascular risk by 30-40%."

# Article schema with E-E-A-T signals
python scripts/schema_generator.py article \
  --title "Understanding Biomarkers" \
  --author "Dr. Sarah Johnson" \
  --credentials "MD, PhD" \
  --date "2025-01-15"

---

Architecture

Phase 1: Analysis (Complete)

Six Python scripts analyze content and generate recommendations:

Script	Purpose	Output
analyze_content.py	Extract metadata, schema, structure	JSON analysis
metadata_validator.py	Validate meta tags, OG, Twitter	Validation report
keyword_analyzer.py	Extract keywords (5 types)	Keyword list
entity_extractor.py	Extract persons, orgs, places	Entity map
schema_generator.py	Generate JSON-LD schemas	Schema files
audit_report.py	Orchestrate analysis, generate reports	JSON, MD, HTML

Phase 2: Implementation (Complete)

Six Python scripts implement optimizations:

Script	Purpose	Impact
content_optimizer.py	Meta descriptions, FAQ, data tables	Structure optimization
platform_optimizer.py	ChatGPT, Perplexity, Claude, Gemini	Platform-specific
voice_optimizer.py	Speakable schema, featured snippets	Voice search ready
freshness_monitor.py	Content age tracking	3.2x citations when fresh
citation_enhancer.py	Statistics, quotation opportunities	+41% and +28% impact
auto_implementer.py	Full pipeline orchestration	Complete automation
indexnow_submit.py	Instant search engine indexing	Minutes vs weeks

Phase 3: Advanced Features (Planned)

• Competitive analysis (SERP top 10 comparison)
• Automated monitoring (SQLite time-series tracking)
• A/B testing framework (statistical significance)
• Analytics integration (Google Search Console, Plausible)

---

Features

Content Analysis

File Types:

• HTML (.html)
• Markdown (.md, .mdx)
• React/JSX (.jsx, .tsx)

Extraction:

• Meta tags (title, description, keywords)
• Open Graph (og:title, og:description, og:image)
• Twitter Cards (twitter:card, twitter:title)
• JSON-LD schema markup
• Content structure (headings, word count, author)

Keyword Analysis

Five keyword types extracted:

1. Primary: Main topic (H1, meta title, URL, first 100 words)
2. Semantic: Related terms (H2/H3, body)
3. LSI: Co-occurring terms (natural language)
4. Long-tail: 3-8 word phrases (FAQ, H3)
5. Question: Who/what/where/when/why/how (FAQ schema)

Keyword Clustering: TF-IDF + cosine similarity groups keywords into semantic clusters for topical authority and pillar content strategy.

Schema Generation

JSON-LD schemas with validation:

• FAQPage: Highest AI citation probability
• Article: E-E-A-T signals (credentials, dates)
• HowTo: Voice search optimized (ISO 8601 durations)
• BreadcrumbList: Site hierarchy
• Organization/LocalBusiness: Entity recognition
• Person: Author profiles with credentials
• Speakable: Voice assistant optimization

Platform Optimization

ChatGPT (40-60% of LLM traffic):

• Authority and credentials (+40% citation boost)
• 1500-2500 words comprehensive coverage
• Primary source citations (PubMed, arXiv)
• Answer-first structure

Perplexity (Freshness-focused):

• Content updated within 30 days (3.2x citations)
• Inline citations with [1], [2] format
• H2→H3→bullets structure (40% more citations)
• Update frequency: 2-3 days (aggressive) or 90 days (minimum)

Claude (Accuracy-focused):

• Primary sources only (91.2% attribution accuracy)
• 5-8 citations with publisher and year
• Transparent methodology
• Acknowledged limitations

Gemini (Community-focused):

• Google Business Profile integration
• User reviews and testimonials
• Local citations (NAP consistency)
• Traditional authority signals

Grokipedia (xAI, launched Oct 2025):

• RAG-based citations (20-30% better factual consistency)
• Transparent version history and licensing
• Primary source attribution (publisher + year)
• Wikipedia-derived content requires CC-BY-SA attribution

Voice Search

• Featured snippet optimization (30-40 words)
• Speakable schema (20-30 second segments)
• FAQ schema (natural language questions)
• Question keyword extraction

Social Media Previews

• Open Graph tags (Facebook, LinkedIn, WhatsApp, Instagram Stories)
• Twitter Cards (summary, summary_large_image)
• Image specifications (1200×630px optimal, 1080×1920px for Instagram Stories)
• Instagram: Limited OG support (Stories only), bio link optimization, 85%+ mobile users
• iMessage optimization

---

Output

Report Location

~/Documents/SEO_Audit_[YYYY-MM-DD]_[HHMM]/
├── audit_report.json          # Structured data
├── audit_report.md            # Markdown report
├── audit_report.html          # Visual dashboard
└── generated_schemas/         # JSON-LD files
    ├── faq_schema.json
    ├── article_schema.json
    └── howto_schema.json

Report Structure

1. Executive Summary: Overall score (0-100), top issues, top wins
2. Metadata Analysis: Meta tags, Open Graph, Twitter Cards, schema
3. Content Structure: Headings, word count, TL;DR, FAQ, author
4. Keyword Analysis: Primary, semantic, LSI, long-tail, question
5. Platform Optimization: ChatGPT, Perplexity, Claude, Gemini recommendations
6. Action Items: Prioritized by criticality (critical, high, medium, low)
7. Generated Assets: Copy-paste ready schema markup

---

Reference Documentation

Located in reference/ directory:

• citation-optimization-guide.md: AI citation strategies (+35-40% boost)
• entity-seo-guide.md: Knowledge Graph optimization
• platform-strategies.md: Platform-specific tactics
• voice-search-guide.md: Voice assistant optimization (29-word answers)
• social-preview-guide.md: Open Graph, Twitter Cards (1200×630px)
• `schema-library.md