SEO & GEO Skills Library

20 skills. 10 commands. Rank in search. Get cited by AI.

English | 中文 | 日本語 | 한국어 | Espanol | Portugues

Claude Skills and Commands for Search Engine Optimization (SEO) and Generative Engine Optimization (GEO). Zero dependencies, works with Claude Code, Cursor, Codex, and 35+ other agents. Content quality scored by the CORE-EEAT Benchmark (80 items). Domain authority scored by CITE Domain Rating (40 items).

SEO gets you ranked in search results. GEO gets you cited by AI systems (ChatGPT, Perplexity, Google AI Overviews). This library covers both.

Why This Library

• 120-item quality frameworks — CORE-EEAT (80 items) + CITE (40 items) with veto gates, not narrative guesswork
• 8 languages, 750+ triggers — EN, ZH, JA, KO, ES, PT with formal, casual, and misspelling variants
• Zero dependencies — pure markdown skills, no Python, no venv, no API keys required
• Tool-agnostic — works standalone or with Ahrefs, Semrush, Google Search Console via MCP
• 6 install methods — ClawHub, skills.sh, Claude Code plugin, git submodule, fork, manual

Quick Start

Works with Claude Code, OpenClaw, Cursor, Codex, and 35+ other agents. No other dependencies.

1. Install — pick the method for your tool:

   | Your tool | Install command | |-----------|----------------| | OpenClaw | clawhub install aaron-he-zhu/<skill-name> — browse all 20 | | Claude Code | /plugin marketplace add aaron-he-zhu/seo-geo-claude-skills (all 20) | | Cursor / Codex / Windsurf / other | npx skills add aaron-he-zhu/seo-geo-claude-skills (all 20) |

   All installation methods remain available at all times. If any marketplace is temporarily unavailable, use an alternative method.

   Install a single skill via skills.sh:

   npx skills add aaron-he-zhu/seo-geo-claude-skills -s keyword-research
   

   Already listed in the table above. Use this method if you need to load locally:

   # Or load locally
   claude --plugin-dir ./seo-geo-claude-skills
   

   Includes marketplace.json, plugin.json, and pre-configured MCP servers. See CONNECTORS.md for MCP setup.

   Add as a submodule for version-pinned updates within an existing project:

   git submodule add https://github.com/aaron-he-zhu/seo-geo-claude-skills.git .claude/skills/seo-geo
   

   Update to the latest version:

   git submodule update --remote .claude/skills/seo-geo
   

   For teams wanting custom modifications:

   1. Fork this repository on GitHub
   2. Clone your fork:

      git clone https://github.com/YOUR-ORG/seo-geo-claude-skills.git
      

   3. Customize skills, add internal connectors, or adjust scoring weights
   4. Install from your fork:

      npx skills add YOUR-ORG/seo-geo-claude-skills
      

   5. Pull upstream updates:

      git remote add upstream https://github.com/aaron-he-zhu/seo-geo-claude-skills.git
      git fetch upstream && git merge upstream/main
      

   git clone https://github.com/aaron-he-zhu/seo-geo-claude-skills.git
   mkdir -p ~/.claude/skills/ && cp -r seo-geo-claude-skills/* ~/.claude/skills/
   

2. Use immediately — no tool integrations required:

   Research keywords for [your topic] and identify high-value opportunities
   

3. Run a command for a one-shot task:

   /seo:audit-page https://example.com/your-page
   

4. Optionally connect tools — edit CONNECTORS.md to map ~~placeholders to your toolstack (Ahrefs, SEMrush, Google Analytics, etc.)

Operating Model

Every skill in this repo now follows the same lightweight contract:

• Trigger: when this skill must be the next move
• Quick Start: shortest successful invocation
• Skill Contract: what the skill reads, writes, and promotes
• Handoff: reusable summary for later skills or later sessions
• Next Best Skill: one primary follow-up, not a long menu

The four cross-cutting skills form the protocol layer:

• content-quality-auditor = publish readiness gate
• domain-authority-auditor = citation trust gate
• entity-optimizer = canonical entity profile
• memory-management = campaign memory loop

Shared references:

• references/skill-contract.md
• references/state-model.md

Automation

Prompt-based hooks run automatically during your session — no configuration needed:

• Session start: loads your project memory and reminds you of open items
• After writing content: recommends a quality audit before publishing
• Session end: offers to save findings for next time

Memory

Three-tier temperature model keeps project context across sessions:

• HOT (80 lines, auto-loaded): goals, hero keywords, active veto items
• WARM (on-demand): audit summaries, research findings, content plans
• COLD (archive): historical data, queried only when needed

Findings promote automatically when referenced frequently. Stale items archive after 90 days.

Where to Begin

| Your Goal | Start Here | Then | |-----------|-----------|------| | Starting from scratch | keyword-research → competitor-analysis | → seo-content-writer | | Write new content | keyword-research | → seo-content-writer + geo-content-optimizer | | Improve existing content | /seo:audit-page <URL> | → content-refresher or seo-content-writer | | Fix technical issues | /seo:check-technical <URL> | → technical-seo-checker | | Assess domain authority | /seo:audit-domain <domain> | → backlink-analyzer | | Full quality assessment | content-quality-auditor + domain-authority-auditor | → 120-item combined report | | Build entity/brand presence | entity-optimizer | → schema-markup-generator + geo-content-optimizer | | Generate performance report | /seo:report <domain> <period> | → periodic monitoring |

Methodology

Skills are organized into four execution phases plus one protocol layer. Use them in order for new projects, or jump to any phase as needed.

 RESEARCH          BUILD            OPTIMIZE          MONITOR
 ─────────         ─────────        ─────────         ─────────
 Keywords          Content          On-Page           Rankings
 Competitors       Meta Tags        Technical         Backlinks
 SERP              Schema           Links             Performance
 Gaps              GEO              Refresh           Alerts

 CROSS-CUTTING / PROTOCOL LAYER ─────────────────────────────────
 Content Quality Gate · Citation Trust Gate · Entity Truth · Memory Loop

Skills

Research — understand your market before creating content

| Skill | What it does | |-------|-------------| | keyword-research | Discover keywords with intent analysis, difficulty scoring, and topic clustering | | competitor-analysis | Analyze competitor SEO/GEO strategies and find their weaknesses | | serp-analysis | Analyze search results and AI answer patterns for target queries | | content-gap-analysis | Find content opportunities your competitors cover but you don't |

Build — create content optimized for search and AI

| Skill | What it does | |-------|-------------| | seo-content-writer | Write search-optimized content with proper structure and keyword placement | | geo-content-optimizer | Make content quotable and citable by AI systems | | [meta-tags-optimizer](https://github.com/aaron-he-zhu/seo-geo-claude-skills/blob/main/build/meta-tags-optimizer