⚠️

重要前提

安装AI Skills的关键前提是：必须科学上网，且开启TUN模式，这一点至关重要，直接决定安装能否顺利完成，在此郑重提醒三遍：科学上网，科学上网，科学上网。查看完整安装教程 →

GEO与AEO优化框架：AI搜索引擎与生成式AI响应的可见性提升方法论

geo-aeo-optimization by schwepps/skills

82 周安装量

10 GitHub Stars

GitHub

安装命令

npx skills add https://github.com/schwepps/skills --skill geo-aeo-optimization

方法论内容创作 SEO

🇨🇳中文介绍

GEO 与 AEO 优化框架

面向人工智能搜索引擎和生成式 AI 响应的可见性专业方法论。

理解 GEO、AEO 与 SEO

SEO  → 在传统搜索结果（蓝色链接）中排名
AEO  → 出现在精选摘要、AI 概览、语音答案中
GEO  → 在生成式 AI 响应（ChatGPT、Claude、Perplexity）中被引用

2025 年现状：

58% 的查询是对话式的
约 47% 的 Google 搜索会出现 AI 概览
由于 AI 答案的出现，传统搜索结果顶部位置的点击率下降了 34.5%
这些策略是互补的，而非相互竞争

何时使用此技能

为 AI 搜索引擎优化内容
提高精选摘要捕获率
审计 AI 引用就绪度
为 AI 可提取性重构内容
实施 E-E-A-T 信号
为语音搜索做准备

三重威胁框架

1. E-E-A-T 基础（所有情况必需）

AI 系统优先考虑可信、权威的来源：

Experience   → 第一手知识、案例研究、原创研究
Expertise    → 资质、知识深度、技术准确性
Authority    → 引用、反向链接、提及、品牌知名度
Trustworthiness → 准确性、透明度、清晰的来源

实施清单：

□ 每篇文章都附有作者简介和资质
□ 链接到权威来源
□ 包含原始数据/研究
□ 展示方法和来源
□ 定期更新内容并显示日期
□ 显示联系信息
□ 包含编辑指南

广告位招租

在这里展示您的产品或服务

触达数万 AI 开发者，精准高效

联系我们

2. AEO：精选摘要与 AI 概览

以答案开头：

前 40-60 个词应直接回答查询
使用问题格式的 H2 标题
提供简洁、可提取的定义

AEO 内容结构：

## 什么是 [主题]？

[1-2 句可直接提取为摘要的直接答案]

[后续是支持性细节和背景]

针对摘要优化的格式：

查询类型	最佳格式
"什么是 X？"	定义段落（40-60 词）
"如何 X"	编号步骤（5-8 步）
"最佳 X"	带简短描述的要点列表
"X 与 Y 对比"	对比表格
"为什么 X？"	直接答案 + 解释

3. GEO：AI 引用优化

C - Clarity    → AI 可以引用的明确陈述
R - Relevance  → 匹配对话式查询意图
A - Authority  → 引用的来源、专家资质
S - Structure  → 可扫描、可提取的格式
P - Precision  → 具体事实、数字、日期

高引用内容元素：

TL;DR 块 置于顶部
带来源的统计数据（"根据 [来源]，X 增长了 Y%"）
带归属的专家引述
关键术语的定义列表
带清晰数据的对比表格
包含完整答案的 FAQ 部分

# [H1: 主要问题或主题]

**TL;DR:** [2-3 句包含关键事实的摘要]

## 什么是 [主题]？ {#definition}

[40-60 词可直接提取的定义 - 针对提取优化]

[带背景的扩展解释]

## [主题] 如何运作 {#how-it-works}

[分步解释]

1.  **第一步**: [描述]
2.  **第二步**: [描述]
3.  **第三步**: [描述]

## 关键统计数据

| 指标 | 数值 | 来源 |
|--------|-------|--------|
| [统计 1] | [数值] | [来源, 年份] |
| [统计 2] | [数值] | [来源, 年份] |

## [主题] 与 [替代方案] 对比

| 因素 | [主题] | [替代方案] |
|--------|---------|---------------|
| [因素 1] | [数值] | [数值] |
| [因素 2] | [数值] | [数值] |

## 常见问题解答

### [问题 1]？

[完整、独立的答案 - 50-100 词]

### [问题 2]？

[完整、独立的答案 - 50-100 词]

## 专家见解

> "[具有具体见解的可引用陈述]"
> — [专家姓名], [职位], [组织]

---
*最后更新: [日期] | 作者: [姓名], [资质]*

事实密度检查清单

Extractability:
□ 顶部有 TL;DR 或摘要
□ 每个段落 <120 词
□ 清晰的主题句（首句 = 主要观点）
□ 用于特性/优点的要点列表
□ 用于对比的表格
□ 带结构化数据的 FAQ 部分

Fact-Density:
□ 引用来源的统计数据
□ 具体的数字和日期
□ 可验证的数据点
□ 带归属的专家引述
□ 原创研究/数据
□ 命名实体（人物、公司、地点）

Structure for AI:
□ 基于问题的 H2 标题
□ 前 40-60 词中的直接答案
□ 逻辑信息层次结构
□ 结构化数据标记（FAQPage、HowTo、Article）

用于 AI 搜索可见性的 robots.txt：

# 允许 搜索/聊天功能
User-agent: OAI-SearchBot
Allow: /

User-agent: ChatGPT-User
Allow: /

User-agent: PerplexityBot
Allow: /

User-agent: ClaudeBot
Allow: /

# 可选：阻止训练（保留搜索）
User-agent: GPTBot
Disallow: /

User-agent: Google-Extended
Disallow: /

关键点： 确保服务器端渲染。AI 爬虫通常不能很好地执行 JavaScript。

面向 AI 的结构化数据标记

优先的结构化数据类型：

FAQPage - 对 AI 引用影响最大
HowTo - 分步可见性
Article + Author - E-E-A-T 信号
Organization - 品牌权威性

FAQPage 示例：

{
  "@context": "https://schema.org",
  "@type": "FAQPage",
  "mainEntity": [{
    "@type": "Question",
    "name": "What is [topic]?",
    "acceptedAnswer": {
      "@type": "Answer",
      "text": "[Complete answer that can be extracted]"
    }
  }]
}

# GEO/AEO 审计报告

**页面:** [URL]
**日期:** [日期]
**总体得分:** X/100

## 执行摘要
[2-3 句 AI 搜索就绪度评估]

## 按类别得分

| 类别 | 得分 | 状态 |
|----------|-------|--------|
| E-E-A-T 信号 | X/25 | [良好/需改进/严重] |
| 内容结构 | X/20 | [良好/需改进/严重] |
| 事实密度 | X/20 | [良好/需改进/严重] |
| 可提取性 | X/20 | [良好/需改进/严重] |
| AI 技术就绪度 | X/15 | [良好/需改进/严重] |

## 优先问题
1.  **[问题]** - [影响] - [修复方法]
2.  **[问题]** - [影响] - [修复方法]
3.  **[问题]** - [影响] - [修复方法]

## AI 可见性评估
[当前引用概率和差距]

## 行动计划

**立即执行:**
- [ ] 为顶部文章添加 TL;DR
- [ ] 实施 FAQ 结构化数据

**短期:**
- [ ] 将标题重构为问题形式
- [ ] 添加作者资质

**持续进行:**
- [ ] 监控 AI 平台引用情况
- [ ] 更新内容新鲜度

快速审计（5 点检查）

TL;DR - 顶部有摘要吗？
结构 - 标题是基于问题的吗？
事实 - 有带来源的统计数据吗？
E-E-A-T - 作者资质可见吗？
结构化数据 - 有 FAQPage/HowTo 标记吗？

90-100: 优秀 - AI 引用概率高
70-89: 良好 - 可能被引用，需小幅改进
50-69: 需改进 - 需要显著优化
低于 50: 严重 - 需要重大重构

2026 年 1 月 24 日

🇺🇸English

GEO & AEO Optimization Framework

Professional methodology for visibility in AI-powered search engines and generative AI responses.

Understanding GEO vs AEO vs SEO

SEO  → Rank in traditional search results (blue links)
AEO  → Appear in featured snippets, AI Overviews, voice answers
GEO  → Get cited in generative AI responses (ChatGPT, Claude, Perplexity)

2025 Reality:

58% of queries are conversational
AI Overviews appear in ~47% of Google searches
Traditional CTR dropped 34.5% for top results due to AI answers
These strategies are complementary, not competing

When to Use This Skill

Optimizing content for AI search engines
Improving featured snippet capture rate
Auditing AI citation readiness
Restructuring content for AI extractability
Implementing E-E-A-T signals
Preparing for voice search

The Triple-Threat Framework

1. E-E-A-T Foundation (Required for All)

AI systems prioritize trusted, authoritative sources:

Experience   → First-hand knowledge, case studies, original research
Expertise    → Credentials, depth of knowledge, technical accuracy
Authority    → Citations, backlinks, mentions, brand recognition
Trustworthiness → Accuracy, transparency, clear sourcing

Implementation Checklist:

□ Author bio with credentials on every article
□ Link to authoritative sources
□ Include original data/research
□ Show methodology and sources
□ Update content regularly with visible dates
□ Display contact information
□ Include editorial guidelines

2. AEO: Featured Snippets & AI Overviews

Lead with the Answer:

First 40-60 words should directly answer the query
Use question-format H2 headers
Provide concise, extractable definitions

Content Structure for AEO:

## What is [Topic]?

[Direct answer in 1-2 sentences that can be extracted as a snippet]

[Supporting details and context follow]

Snippet-Optimized Formats:

Query Type	Best Format
"What is X?"	Definition paragraph (40-60 words)
"How to X"	Numbered steps (5-8 steps)
"Best X"	Bulleted list with brief descriptions
"X vs Y"	Comparison table
"Why X?"	Direct answer + explanation

3. GEO: AI Citation Optimization

The CRASP Framework:

C - Clarity    → Unambiguous statements AI can quote
R - Relevance  → Match conversational query intent
A - Authority  → Cited sources, expert credentials
S - Structure  → Scannable, extractable format
P - Precision  → Specific facts, numbers, dates

High-Citation Content Elements:

TL;DR blocks at the top
Statistics with sources ("According to [Source], X increased by Y%")
Expert quotes with attribution
Definition lists for key terms
Comparison tables with clear data
FAQ sections with complete answers

Content Structure Template

# [H1: Primary Question or Topic]

**TL;DR:** [2-3 sentence summary with key facts]

## What is [Topic]? {#definition}

[Direct definition in 40-60 words - optimized for extraction]

[Expanded explanation with context]

## How [Topic] Works {#how-it-works}

[Step-by-step explanation]

1. **Step One**: [Description]
2. **Step Two**: [Description]
3. **Step Three**: [Description]

## Key Statistics

| Metric | Value | Source |
|--------|-------|--------|
| [Stat 1] | [Value] | [Source, Year] |
| [Stat 2] | [Value] | [Source, Year] |

## [Topic] vs [Alternative]

| Factor | [Topic] | [Alternative] |
|--------|---------|---------------|
| [Factor 1] | [Value] | [Value] |
| [Factor 2] | [Value] | [Value] |

## Frequently Asked Questions

### [Question 1]?

[Complete, standalone answer - 50-100 words]

### [Question 2]?

[Complete, standalone answer - 50-100 words]

## Expert Insights

> "[Quotable statement with specific insight]"
> — [Expert Name], [Title], [Organization]

---
*Last updated: [Date] | Author: [Name], [Credentials]*

Fact-Density Checklist

Extractability:
□ TL;DR or summary present at top
□ Paragraphs <120 words each
□ Clear topic sentences (first sentence = main point)
□ Bullet lists for features/benefits
□ Tables for comparisons
□ FAQ section with schema

Fact-Density:
□ Statistics with sources cited
□ Specific numbers and dates
□ Verifiable data points
□ Expert quotes with attribution
□ Original research/data
□ Named entities (people, companies, places)

Structure for AI:
□ Question-based H2 headers
□ Direct answers in first 40-60 words
□ Logical information hierarchy
□ Schema markup (FAQPage, HowTo, Article)

AI Crawler Configuration

robots.txt for AI Search Visibility:

# ALLOW search/chat features
User-agent: OAI-SearchBot
Allow: /

User-agent: ChatGPT-User
Allow: /

User-agent: PerplexityBot
Allow: /

User-agent: ClaudeBot
Allow: /

# OPTIONAL: Block training (keeps search)
User-agent: GPTBot
Disallow: /

User-agent: Google-Extended
Disallow: /

Critical: Ensure server-side rendering. AI crawlers often don't execute JavaScript well.

Schema Markup for AI

Priority Schema Types:

FAQPage - Highest AI citation impact
HowTo - Step-by-step visibility
Article + Author - E-E-A-T signals
Organization - Brand authority

FAQPage Example:

{
  "@context": "https://schema.org",
  "@type": "FAQPage",
  "mainEntity": [{
    "@type": "Question",
    "name": "What is [topic]?",
    "acceptedAnswer": {
      "@type": "Answer",
      "text": "[Complete answer that can be extracted]"
    }
  }]
}

Output Format

# GEO/AEO Audit Report

**Page:** [URL]
**Date:** [Date]
**Overall Score:** X/100

## Executive Summary
[2-3 sentence AI search readiness assessment]

## Scores by Category

| Category | Score | Status |
|----------|-------|--------|
| E-E-A-T Signals | X/25 | [Good/Needs Work/Critical] |
| Content Structure | X/20 | [Good/Needs Work/Critical] |
| Fact-Density | X/20 | [Good/Needs Work/Critical] |
| Extractability | X/20 | [Good/Needs Work/Critical] |
| AI Technical Readiness | X/15 | [Good/Needs Work/Critical] |

## Priority Issues
1. **[Issue]** - [Impact] - [Fix]
2. **[Issue]** - [Impact] - [Fix]
3. **[Issue]** - [Impact] - [Fix]

## AI Visibility Assessment
[Current citation probability and gaps]

## Action Plan

**Immediate:**
- [ ] Add TL;DR to top articles
- [ ] Implement FAQ schema

**Short-term:**
- [ ] Restructure headers as questions
- [ ] Add author credentials

**Ongoing:**
- [ ] Monitor AI platform citations
- [ ] Update content freshness

Quick Audit (5-Point Check)

TL;DR - Summary present at top?
Structure - Question-based headers?
Facts - Statistics with sources?
E-E-A-T - Author credentials visible?
Schema - FAQPage/HowTo markup?

Scoring Guide

90-100: Excellent - High AI citation probability
70-89: Good - Likely cited with minor improvements
50-69: Needs Work - Significant optimization needed
Below 50: Critical - Major restructuring required

Weekly Installs

Repository

schwepps/skills

GitHub Stars

First Seen

Jan 24, 2026

Security Audits

Gen Agent Trust HubPass SocketPass SnykWarn

Installed on

gemini-cli57

opencode57

codex55

claude-code54

cursor53

github-copilot53

程序化SEO实战指南：大规模创建优质页面，避免内容单薄惩罚

40,500 周安装