geo-content
Skill2 starsChanged 8 days agoMITView on GitHub
Write or retrofit web/blog content so it gets cited by AI search engines — answer-first structure, question headings, sourced statistics, and FAQ schema. Use when creating blog posts, landing/pillar pages, or improving existing pages for AI-search visibility (GEO/AEO).
10 other projects use a version of this.
Adopted by
Most recently taken up by
- alpit-patel/seo-aeo-geo-kittheir own version, first seen Nov 8, 2026
- Gallyan/trouvabletheir own version, first seen Nov 1, 2026
- DoddleHQ/doddle-seo-geo-skillstheir own version, first seen Oct 20, 2026
- impactbrussels/AINativeOStheir own version, first seen Oct 20, 2026
- Ormus-Solutions/LibreGEO-Claude-Codetheir own version, first seen Oct 13, 2026
- techhorizonlabs/thl-opentheir own version, first seen Oct 9, 2026
Used with
Projects with it also use geo.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
ORBWEVASkillgeo
Generative Engine Optimization — optimize content for AI search engines (ChatGPT, Perplexity, Google AI Overviews, Gemini). Use when user wants to rank in AI search, improve AI citability, audit content for LLM visibility, or explicitly invokes /geo commands. Triggers on "GEO", "generative engine", "AI search", "AI overviews", "citability", "Perplexity", "ChatGPT search", or /geo commands.
Also in 4 projects that use this
Other versions10 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 10
- zubair-trabzada/geo-seo-claude★ 11ksame name
- techhorizonlabs/thl-open★ 16same name
- alpit-patel/seo-aeo-geo-kit★ 2same name
- HermeticOrmus/LibreGEO-Claude-Code★ 1same name
- impactbrussels/AINativeOS★ 1same name
- Ormus-Solutions/LibreGEO-Claude-Code★ 1same name
- YogirajA/Claude-Skills★ 1same name
- DoddleHQ/doddle-seo-geo-skills★ 0same name
- Gallyan/trouvable★ 0same name
- kennyolofsson23-netizen/claude-code-config★ 0same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository3 other files it gives its agents.
- geo-auditSkill
- geo-fanoutSkill
- geo-measureSkill
The file
We split this skill into the rules it gives an agent
6 rules. 0 are common, 5 are distinctive, and 1 is very specific to this project.
0
Common
Most of the 9 edited versions we read keep these.
5
Distinctive
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 19
Never invent a statistic or a source.
- Distinctive: line 68
Add genuine EEAT: a real author with credentials.
- Very specific: line 73
Don't add llms.txt, speakable, or keyword-stuffing — none help (folklore).
- Distinctive: line 74
Don't invent statistics, sources, quotes, or facts.
- Distinctive: line 75
Don't produce thin near-duplicate pages at scale — quality per page wins.
- Distinctive: line 80
After writing, confirm the page renders the KeyAnswer, the FAQ, and a
“Common” is measured: kept by at least 6 in 10 of the 9 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
GEO Content
Produce content that AI answer engines cite. The measured winners are
answer-first clarity, Q&A structure, and sourced statistics — bake all three
in every time. Self-contained; if geo-kit's playbook/ and MDX components are
present, use them.
Before writing — gather context
- The target query/intent (mirror how a person would ask it).
- The project's
geo.config(brand,siteUrl,locale,targetIntents) — never hard-code these. - Real facts: names, prices, places, dates, and genuine sourced statistics. Never invent a statistic or a source.
- What does the page have that the things it covers don't? Agents skip a guide and open the venue, product or official page directly unless the guide holds something extra: prices checked on a stated date, verified hours, first-hand notes, a comparison table, the site's own data. Put that in the title and meta description too, since that's what the agent sees before deciding to open the page.
- Year: agents add the current year to most searches unprompted. Keep it in the title of anything time-sensitive, and refresh it with real changes.
The required structure (every page)
- Answer-first block, first thing after the title. 40–70 words that directly
answer the title's implied question — the retrievable answer with the actual
names/prices/places. Use the
<KeyAnswer>component (or a styled lead paragraph). This is the single strongest AI-citation factor. Keep the brand in the page title and the byline with a reviewed date: agents' page readers pick both up. - Body in question-styled sections. Headings phrased as questions where natural ("Where to eat in X?", "Is Y worth it?"). Open each section by answering it, then elaborate. The first sentence must make sense with nothing around it: name the subject instead of "it" or "here", and put the deciding fact (price, dates, distance, open/closed) in that sentence.
- If the site is one of the options, give it the same facts as the others.
An agent's page reader answers the agent's question (price? dates? open in
winter? how to book?) and reports "not stated" for any option that doesn't
say. Its own events, product or service need price, schedule, season and an
absolute booking URL (
https://example.com/events, not/events: the reader drops the domain). Adjectives about yourself get flagged as promotional; facts get repeated. - 1–2 sourced statistics via
<StatCallout value label source sourceUrl>— only real, verifiable numbers with a named source. Omit if none fits; never fabricate. - Closing FAQ via
<Faq>+<FaqItem question answer>(4–5 items) that emits FAQPage JSON-LD. Phrase questions like real search queries; answers must agree with the body. - Freshness: set/update the
updatedAtdate; keep year markers current.
MDX components: geo-kit ships
KeyAnswer,Faq/FaqItem,StatCalloutinsrc/components/mdx/. Note: many MDX pipelines do not evaluate brace-expression attributes (items={[…]}), so FAQ data is authored as nested<FaqItem question="…" answer="…" />children with string props only.
Voice & quality
- Answer the question before elaborating — in the KeyAnswer and each section's first sentence.
- Concrete and specific (named places, real prices) beats generic.
- Match the site's existing tone and locale (
en-GBvsen-USspelling, etc.). - Add genuine EEAT: a real author with credentials; link primary sources.
- Internal-link to related pages and the relevant conversion page.
Do NOT
- Don't add
llms.txt,speakable, or keyword-stuffing — none help (folklore). - Don't invent statistics, sources, quotes, or facts.
- Don't produce thin near-duplicate pages at scale — quality per page wins; raw page count barely correlates with AI visibility.
Verify
After writing, confirm the page renders the KeyAnswer, the FAQ, and a FAQPage JSON-LD script server-side (view-source, not just the client payload), and that every internal link resolves.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

