Imported from reason-machines/marketing-skills (skills/seo-geo-claude-skills/SKILL.md) via skills.sh. Install upstream with npx skills add reason-machines/marketing-skills --skill seo-geo-claude-skills. Copyright stays with the author.

seo-geo-claude-skills

Skill by ara.so — Marketing Skills collection.

Overview

seo-geo-claude-skills is a comprehensive library of 20 specialized skills that enable AI coding agents to perform Search Engine Optimization (SEO) and Generative Engine Optimization (GEO) tasks. Each skill follows a standard contract and includes built-in quality benchmarks: CORE-EEAT (80-item content quality framework) and CITE (40-item domain authority framework).

The library is organized into five phases: Research, Build, Optimize, Monitor, and Cross-cutting protocol skills. All skills work with user-provided data (Tier 1) and optionally integrate with external tools via MCP connectors.

Installation

Claude Code

/plugin marketplace add aaron-he-zhu/seo-geo-claude-skills

ClawHub.ai / OpenClaw

clawhub install aaron-he-zhu/seo-geo-claude-skills
# Or install individual skills:
clawhub install aaron-he-zhu/keyword-research

Generic Agent Skills Hosts (skills.sh)

npx skills add aaron-he-zhu/seo-geo-claude-skills

# Install single skill:
npx skills add aaron-he-zhu/seo-geo-claude-skills -s keyword-research

Gemini CLI

gemini extensions install https://github.com/aaron-he-zhu/seo-geo-claude-skills

Qwen Code

qwen extensions install https://github.com/aaron-he-zhu/seo-geo-claude-skills

Amp

amp skill add aaron-he-zhu/seo-geo-claude-skills

Kimi Code CLI

kimi plugin install https://github.com/aaron-he-zhu/seo-geo-claude-skills.git

Core Commands

The library provides user-facing slash commands for common workflows:

/aaron:auto - Automatically routes to the best skill based on your goal (recommended starting point)
/aaron:max - Maximum-depth exhaustive analysis mode
/aaron:discover - Keyword research workflow
/aaron:compete - Competitor analysis
/aaron:map - SERP and content gap analysis
/aaron:brief - Generate content briefs
/aaron:write - SEO-optimized content creation
/aaron:series - Multi-article content series planning
/aaron:refresh - Update existing content
/aaron:publish - Content quality gate (CORE-EEAT)
/aaron:audit - On-page SEO audit
/aaron:visibility - Schema markup and meta optimization
/aaron:tech - Technical SEO checks
/aaron:authority - Domain authority audit (CITE)
/aaron:watch - Rank tracking
/aaron:report - Performance reporting
/aaron:remember - Memory management (HOT/WARM/COLD)

Maintenance Commands

/aaron:skillify - Review proposed skills or routes
/aaron:evolve - Controlled library changes from evidence
/aaron:guard - Maintainer validation

Skill Structure

Each skill follows this contract:

Quick Start - Immediate activation instructions
Skill Contract - Expected inputs/outputs
Handoff Summary - What to pass to next skill
Data Sources - Where data comes from (user, tools, memory)
Instructions - Step-by-step execution
Reference Materials - Supporting frameworks
Next Best Skill - Recommended workflow progression

The 20 Skills

Research Phase

keyword-research - Discover and prioritize keywords

# Natural language trigger:
"Research keywords for my SaaS product targeting small teams"

# Or direct command:
/aaron:discover "project management software"

competitor-analysis - Analyze competitor SEO strategies

/aaron:compete https://competitor.com

serp-analysis - Analyze search engine results pages

/aaron:map "best project management tools 2024"

content-gap-analysis - Find missing content opportunities

"Find content gaps between my site and competitors for 'team collaboration'"

Build Phase

seo-content-writer - Create SEO-optimized content

/aaron:write "How to improve team productivity with async communication"

geo-content-optimizer - Optimize for AI citation (Perplexity, ChatGPT)

"Optimize this article for AI search engines and citations"

meta-tags-optimizer - Generate optimized meta tags

/aaron:visibility --meta https://example.com/blog/post

schema-markup-generator - Create structured data markup

/aaron:visibility --schema https://example.com/product/widget

Optimize Phase

on-page-seo-auditor - Audit individual pages

/aaron:audit https://example.com/blog/my-article

technical-seo-checker - Check technical SEO issues

/aaron:tech https://example.com

internal-linking-optimizer - Improve internal link structure

"Suggest internal linking improvements for my blog posts about productivity"

content-refresher - Update and improve existing content

/aaron:refresh https://example.com/old-blog-post

Monitor Phase

rank-tracker - Track keyword rankings

/aaron:watch --keywords "project management,team collaboration,async tools"

backlink-analyzer - Analyze backlink profile

"Analyze backlinks for example.com"

performance-reporter - Generate performance reports

/aaron:report --weekly

alert-manager - Configure and manage alerts

"Set up alerts for ranking drops > 3 positions"

Cross-cutting Protocol Skills

content-quality-auditor - 80-item CORE-EEAT quality gate

/aaron:publish https://example.com/draft-article

domain-authority-auditor - 40-item CITE trust assessment

/aaron:authority https://example.com

entity-optimizer - Build canonical entity profiles

"Create entity profile for [Company Name] with knowledge graph optimization"

memory-management - HOT/WARM/COLD project memory + wiki layer

/aaron:remember compile wiki "content strategy"
/aaron:remember query "what keywords did we target last month?"

Working with Memory (v9.9.9+)

The library maintains three memory tiers:

HOT - Active session state (current outputs, open loops)
WARM - Recent project context (skills, routes, evidence)
COLD - Long-term archive (retired files, wiki history)

Wiki Layer Workflow

Phase 1 (Automatic): Index auto-refreshes at memory/wiki/index.md

Phase 2 (Manual Compilation):

# When you have ≥3 WARM files on a topic:
"Compile a wiki page on our content strategy for Q1"
"Synthesize what we know about keyword targeting"

# Handle contradictions when prompted:
# (a) Accept version A
# (b) Accept version B  
# (s) Skip/postpone
# (i) Ignore permanently

Phase 3 (Retirement):

# Preview retirement candidates:
/aaron:guard --wiki --retire-preview

# After confirmation, WARM files move to memory/archive/

# Recover if needed:
/aaron:remember recover wiki
# Or directly:
bash scripts/recover-retired-warm.sh

Clean restart:

rm -rf memory/wiki/  # Reverts cleanly, preserves COLD history

Quality Frameworks

CORE-EEAT (80 items)

Content quality benchmark covering:

Credibility
Originality
Relevance
Expertise
Experience
Authoritativeness
Trustworthiness

Used by: content-quality-auditor, seo-content-writer, geo-content-optimizer

CITE (40 items)

Domain authority benchmark covering:

Credibility signals
Influence metrics
Trust indicators
Expertise markers

Used by: domain-authority-auditor, backlink-analyzer

Common Workflows

Full SEO Campaign

# 1. Research
/aaron:discover "target keyword topic"
/aaron:compete https://competitor.com
/aaron:map "target keyword"

# 2. Build
/aaron:brief "article topic based on research"
/aaron:write "article title from brief"
/aaron:visibility --meta --schema

# 3. Quality gate
/aaron:publish https://example.com/new-article

# 4. Optimize
/aaron:audit https://example.com/new-article
/aaron:tech https://example.com

# 5. Monitor
/aaron:watch --keywords "keyword1,keyword2,keyword3"
/aaron:report --weekly

GEO Optimization

# Optimize content for AI citations (Perplexity, ChatGPT, etc.)
"Optimize this article for generative engine optimization"
/aaron:visibility --geo https://example.com/article

# Entity-first approach:
"Create entity profile for [Brand/Topic]"
"Optimize content hub around [Entity] for AI discoverability"

Content Refresh

/aaron:refresh https://example.com/old-article
# Analyzes current performance
# Suggests content updates
# Identifies new keyword opportunities
# Recommends internal linking improvements

Configuration

Environment Variables

# Optional external tool connectors (see CONNECTORS.md):
export SERP_API_KEY=${SERP_API_KEY}
export AHREFS_API_KEY=${AHREFS_API_KEY}
export SEMRUSH_API_KEY=${SEMRUSH_API_KEY}
export GSC_CREDENTIALS_PATH=${GSC_CREDENTIALS_PATH}

# Memory persistence (default: ./memory):
export AARON_MEMORY_PATH=${AARON_MEMORY_PATH}

All skills work at Tier 1 (user-provided data) without any external API keys.

Connector Tiers

Tier 1: User provides data (copy-paste, screenshots, files)
Tier 2: Fetch-only tools (cURL, Puppeteer, read-only APIs)
Tier 3: Premium APIs with write access (publish, modify, delete)

See CONNECTORS.md for full MCP integration details.

Troubleshooting

Command Not Found

# If /aaron:* commands don't work:
/aaron:auto "your original request"
# The auto command will route to the correct skill

Old /seo:* Commands

# Breaking rename in v9.9.9
# Convert old commands:
/aaron:auto /seo:audit-page https://example.com/blog/post
# Returns: /aaron:audit https://example.com/blog/post

Memory Conflicts

# If wiki contradictions pile up:
/aaron:remember query "contradictions"
# Resolve via (a)/(b)/(s)/(i) prompts

# Nuclear option - clean restart:
rm -rf memory/wiki/
# Next session rebuilds from WARM

Skill Routing Issues

# Review available skills and routes:
/aaron:skillify

# Check current memory state:
/aaron:remember query "last 5 skills used"

# Validate library health:
/aaron:guard --validate

Advanced Usage

Custom Skill Chaining

# Explicit skill sequence:
"Use keyword-research, then competitor-analysis, then seo-content-writer for 'project management automation'"

Batch Operations

# Audit multiple pages:
/aaron:audit https://example.com/page1
/aaron:audit https://example.com/page2
/aaron:audit https://example.com/page3

# Or natural language:
"Audit all pages in sitemap.xml for on-page SEO"

Report Customization

/aaron:report --weekly --format=json --output=reports/
/aaron:report --monthly --include-competitors --email

Best Practices

Start with /aaron:auto - Let the library route your request automatically
Use memory-management - Compile wiki pages once you have recurring topics
Quality gate everything - Run /aaron:publish before publishing content
Monitor continuously - Set up /aaron:watch alerts for key metrics
Document decisions - Memory layer captures why you chose specific strategies
Combine auditors - Use both content-quality-auditor (80 items) + domain-authority-auditor (40 items) for complete assessment

Files and Directory Structure

seo-geo-claude-skills/
├── commands/              # Slash command definitions
├── research/             # Research phase skills
│   ├── keyword-research/
│   ├── competitor-analysis/
│   ├── serp-analysis/
│   └── content-gap-analysis/
├── build/                # Build phase skills
│   ├── seo-content-writer/
│   ├── geo-content-optimizer/
│   ├── meta-tags-optimizer/
│   └── schema-markup-generator/
├── optimize/             # Optimize phase skills
│   ├── on-page-seo-auditor/
│   ├── technical-seo-checker/
│   ├── internal-linking-optimizer/
│   └── content-refresher/
├── monitor/              # Monitor phase skills
│   ├── rank-tracker/
│   ├── backlink-analyzer/
│   ├── performance-reporter/
│   └── alert-manager/
├── cross-cutting/        # Protocol skills
│   ├── content-quality-auditor/
│   ├── domain-authority-auditor/
│   ├── entity-optimizer/
│   └── memory-management/
├── references/           # Frameworks and benchmarks
│   ├── core-eeat-benchmark.md
│   ├── cite-domain-rating.md
│   ├── auditor-runbook.md
│   └── skill-contract.md
├── scripts/              # Helper scripts
│   └── recover-retired-warm.sh
└── memory/               # Memory persistence
    ├── wiki/
    ├── archive/
    └── index.md

License

Apache License 2.0

About