Preview
What it is.
It starts by making you check your priority queries in ChatGPT, Perplexity and Google AI Overviews and record who gets named. Everything after that is planned against the baseline you just took.
It separates two goals people conflate, being cited as a source and appearing inside an AI Overview for a query, because those need different work. Content types, existing search strength and structured data all feed the plan, and llms.txt is covered with the argument for it.
What you get.
- A baseline check across the assistants for your priority queries
- The two goals separated: citation as a source, and AI Overview presence
- Content-type guidance, since docs and comparison pages behave differently
- Structured data recommendations tied to the citation goal
- llms.txt and Open Knowledge Format guidance, with the argument for each
How to set it up.
Install with npx skills add coreyhaines31/marketingskills --skill ai-seo -a claude-code.
List the queries that matter commercially before anything else. This is the measurement set you will keep re-checking.
Run those queries manually through ChatGPT, Perplexity and AI Overviews and record who gets named. That is the baseline.
Say which goal you want. Getting cited as a source and appearing in AI Overviews for a query need different work.
Bring your content types into it. Comparison pages and documentation get cited on different query shapes than blog posts.
Check your structured data next, because it is one of the levers the skill treats as concrete.
Pricing Plans
Free. MIT license.
Checked against the repo in July 2026.
Use cases
Taking the first baseline
Leadership asks how you show up in AI answers and nobody knows. List the queries that matter commercially, run them through the assistants by hand, and record who gets named; that record is what every later change gets measured against.
Splitting the two goals
The team says AEO and means two different things. State whether you want to be cited as a source or to appear inside an AI Overview; the plan comes back different because the work behind each is.
Fitting AEO to strong SEO
Your traditional rankings are already good. Bring that search strength into the run along with your content types; the recommendations narrow to what's worth adding, including whether llms.txt earns a place.
The monthly re-check
One check tells you what a single assistant said on one day. Re-run the same query set each month; movement against that first record is the signal you report.
Best for
The AEO question with no baseline
It makes you measure before you act, which stops a quarter of content work aimed at a problem you never confirmed.
Deciding between citation and Overview presence
These are different goals with different work behind them, and the skill separates them explicitly.
A team with strong traditional SEO
It asks for your existing search strength, because that changes what AI-search work is worth doing.