Ai search optimization
Curated SEO Agent Skills pack for AI agents (Agent Skills open standard): technical audit, content briefs, JSON-LD schema, GEO/AEO, and anti-spam quality gates.
npx -y skills add gaguero/seo-agent-skills --skill ai-search-optimizationAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
What its author says it does
Copied from the file, not written here
Optimizes content for AI-driven search and answer engines — Google AI Overviews, ChatGPT search, Perplexity, Gemini, Copilot — also called GEO (Generative Engine Optimization) and AEO (Answer Engine Optimization). Makes content extractable and citable: direct up-front answers, clear definitions, comparison tables, genuine FAQs, named entities, and verifiable external authority. Use when someone wants to appear or be cited in AI answers, asks about GEO/AEO, AI Overviews, "ranking in ChatGPT/Perplexity", LLM/answer-engine visibility, or zero-click search.
The file declares its own license as MIT. That is the author’s claim about this one file, and it is not the same thing as the license GitHub reports for the repository, which is listed with the other numbers below.
SKILL.md
4.2 KB, 847 tokens by cl100k_base, as published. Nobody here has run it
AI Search Optimization (GEO / AEO)
Answer engines synthesize a response and cite a handful of sources. Your goal is
to be one of those sources. Crucially, the set of cited pages does not perfectly
match the classic organic top 10 — so being citable is its own discipline on top
of ranking well. This skill assumes the page already passes technical and content
fundamentals (route gaps back to seo-technical-audit / seo-content-brief).
How answer engines pick sources (working model)
They favor content that is:
- Extractable — the answer is stated plainly, early, in self-contained units.
- Trustworthy — clear authorship, citations, accuracy, real expertise (E-E-A-T).
- Specific & current — concrete facts, numbers, dates; not vague filler.
- Well-structured — headings, lists, tables, Q&A that map to a sub-question.
- Corroborated — entities and claims that match what other authoritative sources say (consensus and external mentions matter).
- Accessible to crawlers — not blocked to AI crawlers you want to allow; key content in the initial HTML, not JS-only.
What to do to a page
1. Lead with the answer
For each key question, put a direct, complete answer in the first 1–2 sentences of its section, then expand. Answer-engine extraction grabs the concise statement; burying it loses the citation.
2. Structure for extraction
- One clear H2 per sub-question, phrased the way people ask it.
- Use definition sentences ("X is …"), comparison tables, step lists,
and genuine FAQs (only real Q&A — coordinate with
seo-schema-builder). - Keep paragraphs tight and self-contained so a snippet stands alone.
3. Make entities & facts unambiguous
- Name people, places, products, organizations explicitly (no vague "we"/"it").
- State facts with numbers, dates, and units. Attribute and cite primary sources — engines reward verifiable claims and may follow citations.
4. Demonstrate authority
- Real, credentialed author + bio. About/contact transparency.
- Earned external mentions/links and consistent entity info across the web (NAP, brand name, sameAs) raise the odds of being treated as authoritative.
5. Add original value
Answer engines compress the obvious; they cite sources that add something: first-hand experience, original data/tests, expert judgment, local specifics. Generic restated content rarely earns a citation.
6. Manage crawler access deliberately
Decide which AI crawlers you allow (e.g. Googlebot for AI Overviews vs.
GPTBot, PerplexityBot, ClaudeBot, Google-Extended) and set robots.txt
intentionally. Allowing them is a prerequisite to being cited by that engine;
blocking is a valid business choice but forfeits that visibility.
What NOT to do
- Don't fabricate FAQs, stats, quotes, or sources to look citable — answer engines
and Google's spam systems penalize this, and it's dishonest. (Hard-blocked by
seo-quality-gates.) - Don't keyword-stuff "for the LLM". Write for the human; structure for extraction.
- Don't assume AEO replaces SEO — it's additive. The same index and trust signals feed both.
Output
An annotated optimization plan or revised page: the up-front answers, the
extraction-friendly structure, entities/facts to make explicit, citations to add,
and crawler-access recommendations. Genuine FAQ/structured data → seo-schema-builder;
final check → seo-quality-gates.
See references/answer-engines.md for engine-specific notes and a measurement approach.
What ships with it: 2 files
3.8 KB alongside SKILL.md
references/
- answer-engines.md2.5 KB
- contract.yaml1.2 KB