_{English · 简体中文}

gefei-seo-skill

An Agent Skill for Google SEO in Practice — a knowledge base that AI coding assistants (Claude Code, Codex, OpenCode, etc.) can load and query.

Install

npx skills add ifyour/gefei-seo-skill

Or manually: clone this repo and drop the gefei-seo/ directory into your skills folder (~/.agents/skills/, ~/.claude/skills/, or a project-level .claude/skills/ all work).

Knowledge structure

File	Contents
gefei-seo/SKILL.md	Core framework + chapter/topic index (entry point)
gefei-seo/chapters/	22 chapter digests, loaded on demand
gefei-seo/glossary.md	Key terms
gefei-seo/patterns.md	11 reusable patterns
gefei-seo/cheatsheet.md	Decision quick reference (rules / hard numbers / decision trees)

Usage

After installing, the agent invokes the skill automatically whenever you discuss SEO. Three levels of loading granularity:

• No argument → loads the core framework (the three-sentence ranking model, KGR, publishing cadence, etc.)
• By topic → KGR, link building, programmatic SEO, penalty recovery …, reads the matching chapter
• By chapter → ch09, ch16 …, loads a chapter file directly
• Browse → ask "what chapters are there" to see the index

No commands to memorize — just ask in natural language, for example:

Use KGR to judge whether this keyword is worth targeting
How should a new site build links in its first three months
Is this tool site a good fit for programmatic SEO

Use cases

Keyword research

• Evaluate whether a keyword is worth pursuing: search volume, competition, KGR judgment
• Discovering and prioritizing emerging / trending keywords
• Building long-tail keyword matrices

Site building & on-page optimization

• Page structure and TDH (Title/Description/Heading) for new sites
• SSR/SSG choices, meeting Core Web Vitals
• How content sites differ from tool sites

Links & growth

• Link-building cadence and safety boundaries (avoiding penalties)
• Programmatic SEO: when it fits and how to avoid doorway pages
• Multilingual SEO and hreflang setup

AI search era

• Getting cited in AI Overviews / AI search results
• Day-to-day GSC/GA analysis and anomaly triage

Firefighting

• Troubleshooting a sudden traffic drop
• Diagnosing and recovering from a Google penalty

License & data source

• Code and curated content in this repo are released under MIT.
• Knowledge content comes from Jayin/gefei-seo-cookbook.
• Numbers in the book (weight percentages, costs, etc.) reflect the time of writing and may be outdated — defer to official Google docs and live data.