Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Optimize content and site structure for AI-driven search experiences including AI overviews, large language model citations, generative answer engines, and AI assistants. Use this skill whenever the user wants to optimize for AI search, get cited by language models, appear in AI overviews, build llms.txt, structure content for AI extraction, or future-proof their SEO for the shift from blue links to AI answers. Triggers on AEO, GEO, AI search, AI SEO, AI overview, generative search, LLM optimiza
.claude/skills/rampstackco-seo-aeo-geo/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-02 | ✗→✓ | ▲ Improved | 23% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 22% | 0% |
| case-21 | ✗→✓ | ▲ Improved | 28% | 0% |
| case-04 | ✓→✓ | ＝ Same ✓ | 50% | 0% |
| case-06 | ✓→✓ | ＝ Same ✓ | 80% | 0% |
Answer Engine Optimization (AEO) and Generative Engine Optimization (GEO). Make content discoverable, extractable, and citable by AI search experiences.
This skill encodes principles. AI search products evolve fast. The principles age slower than the products.
seo-onpage or seo-technical)seo-keyword)seo-offpage)This skill stacks on top of those. Strong AEO/GEO requires strong fundamental SEO underneath.
AI search visibility comes from five stacked layers. Each layer compounds.
AI systems extract facts and pull citations from content. Make extraction easy.
AI cites sources it considers authoritative. Earn that consideration.
Schema is how you speak machine-readable language. AI assistants parse it heavily.
sameAs links to verifiable profilesBeyond traditional SEO, AI tools need access patterns of their own.
/llms.txt describing the site's content, key URLs, and what topics the site covers. See references/llms-txt-guide.md.article, section, nav, main) help AI parse structure.AI builds knowledge graphs and prefers entities with multiple consistent signals.
sameAs propertiesDefault output is a markdown plan at aeo-geo-strategy.md. Structure:
This skill's output depends on data, measurements, or tool results it cannot generate on its own. When a required input, tool, or data source is unavailable or unverifiable, the sanctioned output is the deliverable with the gap stated: what was needed, what was actually obtained or verified, and which parts of the output are affected. Fabricating, estimating, or interpolating a required number to complete the deliverable is never sanctioned. A stated gap is a complete answer.
references/llms-txt-guide.md - How to write a useful llms.txt, with examples.references/extraction-friendly-patterns.md - Content patterns that AI extracts cleanly, with before/after examples.| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 36,237 | 34,059 | -6% | 1 | 1 | 0% | 6,272 | 8,069 | +29% | 0 | 0 | — |
case-02 | fail→pass | 34,651 | 31,118 | -10% | 1 | 1 | 0% | 6,246 | 7,674 | +23% | 0 | 0 | — |
case-03 | fail→pass | 34,365 | 31,056 | -10% | 1 | 1 | 0% | 6,250 | 7,611 | +22% | 0 | 0 | — |
case-04 | pass→pass | 15,786 | 14,592 | -8% | 1 | 1 | 0% | 2,916 | 4,360 | +50% | 0 | 0 | — |
case-05 | fail→fail | 9,649 | 12,542 | +30% | 1 | 1 | 0% | 1,706 | 3,968 | +133% | 0 | 0 | — |
case-06 | pass→pass | 18,126 | 21,375 | +18% | 1 | 1 | 0% | 2,987 | 5,367 | +80% | 0 | 0 | — |
case-07 | pass→pass | 12,736 | 13,540 | +6% | 1 | 1 | 0% | 2,100 | 4,111 | +96% | 0 | 0 | — |
case-08 | pass→pass | 26,272 | 23,794 | -9% | 1 | 1 | 0% | 4,817 | 6,123 | +27% | 0 | 0 | — |
case-09 | pass→pass | 12,815 | 7,939 | -38% | 1 | 1 | 0% | 1,957 | 3,067 | +57% | 0 | 0 | — |
case-10 | pass→pass | 14,168 | 9,733 | -31% | 1 | 1 | 0% | 2,254 | 3,589 | +59% | 0 | 0 | — |
case-11 | pass→pass | 10,077 | 6,499 | -36% | 1 | 1 | 0% | 1,680 | 2,813 | +67% | 0 | 0 | — |
case-12 | pass→pass | 16,394 | 13,581 | -17% | 1 | 1 | 0% | 2,779 | 4,473 | +61% | 0 | 0 | — |
case-13 | pass→pass | 17,652 | 14,823 | -16% | 1 | 1 | 0% | 3,167 | 4,270 | +35% | 0 | 0 | — |
case-14 | pass→pass | 5,302 | 4,879 | -8% | 1 | 1 | 0% | 973 | 2,706 | +178% | 0 | 0 | — |
case-15 | pass→pass | 13,682 | 9,041 | -34% | 1 | 1 | 0% | 2,336 | 3,316 | +42% | 0 | 0 | — |
case-16 | pass→pass | 7,202 | 4,774 | -34% | 1 | 1 | 0% | 1,384 | 2,736 | +98% | 0 | 0 | — |
case-17 | pass→pass | 14,132 | 9,947 | -30% | 1 | 1 | 0% | 2,297 | 3,592 | +56% | 0 | 0 | — |
case-18 | pass→pass | 16,003 | 8,939 | -44% | 1 | 1 | 0% | 2,318 | 3,183 | +37% | 0 | 0 | — |
case-19 | pass→pass | 17,883 | 11,792 | -34% | 1 | 1 | 0% | 2,545 | 3,562 | +40% | 0 | 0 | — |
case-20 | pass→pass | 16,475 | 9,428 | -43% | 1 | 1 | 0% | 2,499 | 3,222 | +29% | 0 | 0 | — |
case-21 | fail→pass | 13,969 | 5,248 | -62% | 1 | 1 | 0% | 1,991 | 2,555 | +28% | 0 | 0 | — |
case-22 | pass→pass | 15,597 | 11,570 | -26% | 1 | 1 | 0% | 2,246 | 3,508 | +56% | 0 | 0 | — |
case-23 | pass→pass | 13,553 | 9,941 | -27% | 1 | 1 | 0% | 1,961 | 3,279 | +67% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 23 cases were attempted. The headline lift of +13 percentage points is the difference between those two pass rates over the 23 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.