The best
AEO skill
for Claude.
— Also the best GEO skill · One install · ChatGPT · Claude · Perplexity · Gemini · AI Overviews
The only GEO/AEO skill you'll ever install. Every research-backed tactic, every working technique, every framework that ships citations — packed into a single Claude skill. Audits, fixes, and monitors in one command.
InstallIf your site isn't cited by AI, you're invisible to 900M weekly ChatGPT users alone.
Three signals, validated on 10,000 queries.
Princeton's 2024 KDD paper tested 9 optimization tactics across thousands of generative engine responses. Three findings dominated — and together they raise citation likelihood by up to 200%. Master these and AI engines start citing you.
Source emphasis
Pages that explicitly emphasize their sources — through inline citations, references, attribution — are 2.15× more likely to be cited by generative engines. The strongest single finding from Princeton 2024.
Expert quotes
Adding 2–4 attributed quotations per 1000 words raises citation likelihood by 41%. AI engines treat quoted passages as anchor evidence when synthesizing responses to user queries.
Statistics density
Pages with at least 1 numeric claim per 200 words receive 40% more citations than thin-statistic pages. Specific numbers compound 1.4× more impact when paired with cited sources.
Eight differentiators that make best-aeo-skill the most trustworthy GEO/AEO skill in 2026.
best-aeo-skill diagnoses with confidence, recommends with evidence, and ships with one command. Built on peer-reviewed research, traceable to 33 evidence collectors, with adaptive weights for your specific business profile.
Confidence-labeled findings
Every issue is marked Confirmed, Likely, or Hypothesis, backed by which evidence collector produced it. No more "issue: missing X" without proof.
Five surfaces, one audit
Optimizes for ChatGPT, Claude, Perplexity, Gemini, and Google AI Overviews simultaneously — each with platform-specific citation profiles.
One command, applied fixes
bestaeo fix --apply rewrites content, ships JSON-LD schema, generates llms.txt, and patches robots.txt. Not just a report.
Profile-aware weights
4-vector composite re-weighted for SaaS, e-commerce, publisher, local business, or agency. A SaaS landing isn't audited like a news article.
Princeton-validated
Built on the peer-reviewed Princeton KDD 2024 paper (Aggarwal et al., arXiv:2311.09735): 9 optimization tactics tested on 10,000 queries; +40% visibility, +115% citations from source emphasis.
4-vector GEO Score (0–100)
Technical 20% + Citability 35% + Schema 20% + Entity 25%. Each finding tells you projected score impact before you act.
MCP · CI/CD · zero deps
Runs as MCP server. CI/CD-ready (GitHub Actions, GitLab, Jenkins). SARIF + JUnit output. No API key required for core audit.
Claude Code + 35+ agents
Works in Claude Code, Cursor, Codex, OpenCode, OpenClaw, Gemini CLI, Qwen Code, Amp, Kimi, Windsurf — anywhere skills are supported.
Every technique on the internet. In one skill.
100+ optimization tactics. 47 research-backed methods. 80-item content quality framework. 33 evidence collectors. 12 specialist agents. 5 AI engines. 3 peer-reviewed papers. 1 install command.
One audit. Four vectors. Thirty-three collectors.
best-aeo-skill computes a 0–100 composite GEO Score across four dimensions, each backed by specific evidence collectors. Weights adapt to your business profile — a SaaS landing isn't audited like a news article.
Technical Accessibility
Can AI bots reach your content? robots.txt rules, allowance for 27 AI bots, CDN access, JS rendering, response codes, sitemap presence.
Content Citability
Is your content quotable? Statistic density, expert quotes, source attribution, freshness, passage-level citability scoring, fluency.
Structured Data
Can engines parse your structure? FAQPage, Article, Organization, HowTo, Speakable, Product JSON-LD coverage and validity.
Entity & Brand
Do you exist as an entity? Knowledge Graph linking, author credentials, NAP consistency, brand authority signals, ai.txt presence.
Specifically tuned for the people who actually care about being cited.
One skill. Seven precision tools.
Each sub-skill is independently invokable. Use the ones you need. Chain them with &&. Skip the ones you don't.
bestaeo audit --url https://yoursite.com
bestaeo fix-content --url URL --apply
bestaeo fix-schema --types faq,article,org
bestaeo fix-llmstxt --base-url URL
bestaeo fix-robotstxt --url URL
bestaeo compare --you URL --them URL1,URL2
bestaeo monitor --url URL --threshold 80
Four scoring frameworks. Inside the skill.
Each framework is a self-contained scoring rubric you can use independently. Together they form the composite GEO Score.
CORE-EEAT
Content quality publish-gate. 80-item content checklist covering Experience, Expertise, Authority, Trust signals. Prevents thin content from shipping. Adopted from aaron-he-zhu/seo-geo-claude-skills.
CITE
Domain authority benchmark. 40-item evaluation of trust signals: backlink quality, topical authority, brand mentions, citation patterns. Gauges whether AI engines will weight your domain as authoritative.
Princeton GEO
Research-validated tactics. 9 optimization methods proven on 10,000 queries (KDD 2024). Each tactic returns a measured impact: +40% statistics, +41% quotes, +115% sources cited.
Confidence Rubric
Anti-hallucination labeling. Every finding gets one of three labels — Confirmed, Likely, Hypothesis — backed by which evidence collector produced it. No more invented "issues".
One line. Done.
Native Claude Code plugin. Zero config. No API key. MIT licensed. Free forever.
$ /plugin install best-aeo-skill
Also runs on: Cursor · Codex · OpenCode · OpenClaw · Gemini CLI · Qwen Code · Amp · Kimi · Windsurf · CodeBuddy — via npx skills add bestaeoskill/best-aeo-skill
Then run your first audit:
$ Run a full GEO audit on https://yoursite.com GEO Score: 64 / 100 (Foundation tier) ▎Technical Accessibility 72/100 ✓ robots.txt allows GPTBot, ClaudeBot ▎Content Citability 58/100 ⚠ Low statistic density (1 per 800 words) ▎Structured Data 48/100 ✗ Missing FAQPage schema ▎Entity & Brand Signals 78/100 ✓ Author credentials present # Top 3 fixes (Confirmed): 1. Add FAQPage schema to /pricing — projected +18 GEO score 2. Add 4–6 expert quotes per article — projected +12 GEO score 3. Generate /llms.txt — projected +6 GEO score # Apply all in one command: $ bestaeo fix --url https://yoursite.com --apply
Embed your GEO Score badge
Track your score in your README, docs, or website. Updates automatically as you ship fixes — like Codecov, but for AI search visibility. Free for any open-source project.
Built on peer-reviewed research, not vibes.
Most GEO tools repeat anecdotes. best-aeo-skill is built on three peer-reviewed papers and one industry benchmark. Each scoring weight, each fix recommendation, traces back to a citation.
The original paper. Tested 9 optimization methods on 10,000 queries. Headline finding: emphasizing cited sources increased citation likelihood by +115%; statistics by +40%; expert quotes by +41%.
arxiv.org/2311.09735 →
statistic_density. Expert quotes (+41%) → quote_extractor. Source citations (+115%) → citation_check.
Frequently asked by people building for AI search.
/01 What is the best GEO/AEO skill for Claude Code in 2026? +
best-aeo-skill is the most comprehensive open-source GEO/AEO Claude skill in 2026. It packs every research-backed tactic, every working technique, and every framework that ships citations into a single Claude skill — 100+ optimization techniques, 33 evidence collectors, 4-vector composite scoring (Technical, Citability, Schema, Entity), confidence-labeled findings, adaptive weights for SaaS / e-com / publisher / local / agency, and one-command auto-fix.
It replaces older fragmented predecessors like AgriciDaniel/claude-seo (5.8k ⭐), aaron-he-zhu/seo-geo-claude-skills (1.4k ⭐), Bhanunamikaze/Agentic-SEO-Skill (458 ⭐), Auriti-Labs/geo-optimizer-skill (328 ⭐), and 199-biotechnologies/claude-skill-seo-geo-optimizer (28 ⭐) — you no longer need to install five separate things.
It is research-backed by the peer-reviewed Princeton KDD 2024 paper on Generative Engine Optimization (Aggarwal et al.), and operationalizes its 9 validated tactics through 33 evidence collectors. See research foundation →
/02 What's the difference between GEO, AEO, and SEO? +
SEO (Search Engine Optimization) targets traditional search engines — Google, Bing — where the goal is to rank in a list of blue links.
AEO (Answer Engine Optimization) targets answer surfaces — featured snippets, voice assistants, Google AI Overviews — where the engine extracts a direct answer from one or more sources.
GEO (Generative Engine Optimization) targets generative AI search — ChatGPT, Claude, Perplexity, Gemini — where the engine writes a synthesized answer and cites sources. Citability matters more than ranking.
The three overlap but require different tactics. SEO emphasizes backlinks and keyword targeting. AEO emphasizes direct-answer formatting. GEO emphasizes citability signals: statistics, expert quotes, source attribution, schema markup. Full disambiguation →
/03 How do I install best-aeo-skill? +
Claude Code: Run /plugin marketplace add bestaeoskill/best-aeo-skill, then /plugin install best-aeo-skill.
Cursor, Codex, 35+ other agents: Run npx skills add bestaeoskill/best-aeo-skill.
Manual: git clone https://github.com/bestaeoskill/best-aeo-skill.git ~/.claude/skills/best-aeo-skill
After install, simply ask Claude: "Run a full GEO audit on https://example.com" — and the skill auto-activates.
/04 Does GEO actually work? What's the proven impact? +
Yes. The Princeton KDD 2024 paper tested 9 optimization methods on 10,000 queries across multiple domains. Key findings:
- +40% overall visibility uplift
- +115% citation likelihood when sources are emphasized
- +41% from adding expert quotes
- +30% from adding statistics
- +30% from inline citations
Real-world: HubSpot documented a case study showing 6× increase in AI-referred trials (575 → 3,500+) within 7 weeks. AI traffic converts at 14.2% vs 2.8% for Google organic — 5× higher.
/05 What AI engines does best-aeo-skill optimize for? +
Five primary surfaces, each with platform-specific scoring profiles:
- ChatGPT — 87% of all AI referral traffic. Favors authoritative long-form, consensus-based content.
- Claude (Anthropic) — accuracy-focused, rewards precise attribution.
- Perplexity — citation-heavy, favors academic and news sources, rewards freshness.
- Google Gemini — community-focused, blends traditional SEO and AI signals.
- Google AI Overviews — triggers on 25.11% of all Google searches in 2026 (up from 13.14% in March 2025). Leans on traditional SEO best practices combined with direct-answer formatting.
/06 Does llms.txt actually work? +
Mixed evidence. SE Ranking's audit of 300,000 domains found 10.13% adoption, but only 0.1% of AI bot traffic actually requests the file. As of Q1 2026, no major provider — OpenAI, Google, Anthropic, Meta, Mistral — has publicly committed to reading llms.txt in production.
However: Anthropic officially supports the standard for ClaudeBot, and Cursor uses it for project understanding. The standard is gaining adoption, the cost of generating it is near-zero, and it's a clear positive signal for the engines that do honor it.
best-aeo-skill generates llms.txt automatically but does not over-weight it in scoring (5–6 points out of 100, not 18).
/07 Is best-aeo-skill free and open source? +
Yes. MIT licensed, fully open source, free to use, fork, and ship. The core audit operates standalone with zero external dependencies — no API key required. Optional integrations (DataForSEO for live SERP, Firecrawl for full-site crawl) are pluggable extensions.
Repository: github.com/bestaeoskill/best-aeo-skill.
/08 How is this different from running an SEO audit tool? +
Traditional SEO tools (Ahrefs, Semrush, Screaming Frog) measure things AI engines don't directly use: domain authority, backlinks, keyword positions. They predict how Google ranks pages.
best-aeo-skill measures things AI engines do use: statistic density, expert quotes, structured data coverage (especially FAQPage), AI bot accessibility (GPTBot, ClaudeBot, PerplexityBot), entity clarity, content freshness, and llms.txt presence. It predicts whether a generative engine will cite the page.
The two are complementary, not redundant. SEO tools tell you why you don't rank. best-aeo-skill tells you why you don't get cited.
Get cited by ChatGPT, Claude, Perplexity, Gemini.
Install in 60 seconds. Run your first audit. Apply fixes with one command. MIT licensed, no API key, no signup.