Skip to content
Agentmart
Submit a tool

on-page-seo-aeo-optimization

Turn a draft or existing page into something both Google *and* AI answer engines can index, rank, and cite.

Skill by narayan-metaflow

IndexedMIT★ 15
Write a review

Chalk Labs

Need help deploying on-page-seo-aeo-optimization?

Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.

Book a call

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/narayan-metaflow-on-page-seo-aeo-optimization

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

Turn a draft or existing page into something both Google *and* AI answer engines can index, rank, and cite. This skill operates at the single-page level — keyword research and strategy are inputs, not outputs.

Use cases: marketing, research

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.

Tools
No tools found
Secrets
None asked for
Install scripts
None found

Network hosts it talks to 1

  • codeload.github.com · found in install command

Files it touches

  • ~/.claude/skills/on-page-seo-aeo-optimization

What's inside

Skill name
on-page-seo-aeo-optimization
Description
Turn a draft or existing page into something both Google *and* AI answer engines can index, rank, and cite. This skill operates at the single-page level — keyword research and strategy are inputs, not outputs.
License
MIT
Instructions
259 lines in SKILL.md · view on GitHub

Files (1)

  • SKILL.md

SKILL.md (first lines)

## Why this matters

- The "ski ramp" is the single most important on-page pattern for AEO: **44.2% of LLM citations come from the first 30% of the page** (large-sample citation analysis, ~1.2M citations, p-value 0.0).
- **78.4% of citations containing questions come from H2 headings** — LLMs treat H2s as prompts. If your H2s aren't phrased as questions users would ask, you leave citation opportunity on the table.
- **AI crawlers do not execute JavaScript.** Any critical content that renders client-side is invisible to ChatGPT, Perplexity, and Claude. The on-page check must include "does this render in raw HTML?"
- Traditional on-page factors (title tags, schema, internal linking) remain load-bearing for Google — roughly the same rules as the last decade, with the AEO layer added on top.

## Process

1. **Verify target keyword and intent alignment.** Read the brief (from seo-content-strategy). Confirm the target keyword and intent match the SERP reality. If the SERP has shifted since the brief was written, adjust before optimizing.
2. **Optimize title tag and meta description.** Title is the single highest-leverage on-page element.
3. **Structure header hierarchy.** H1 is the topic. H2s are question-based subtopics. H3s are supporting detail. Apply the ski ramp.
4. **Optimize body content.** Front-load the answer. Definitive language. Entity density. Raw-HTML-visible content.
5. **Implement internal links.** Every page links to its pillar. Pillars link to all clusters. Clusters cross-link 2-3 related clusters.
6. **Optimize images.** Alt text, dimensions, modern formats, lazy loading.
7. **Implement schema markup (JSON-LD).** Validate with Rich Results Test before publish.
8. **Run the pre-publish checklist.** Confirm every ski ramp and technical element is in place.

## Frameworks

### Title tag formula

```
[Primary keyword] + [Specificity hook] + [Brand]
```

- **Primary keyword up front** (within the first 60 characters; fronted where natural).
- **Specificity hook** = year, number, format, or differentiator. Examples: "for 2026," "[5 Methods]," "Complete Guide," "with Examples."
- **Brand** at the end, separated by `|` or `–`. Skippable for very long titles.
- **Length target:** 50-60 characters. Google truncates around 580px pixel width; 60 chars is a safe proxy.
- **Uniqueness:** Every title on the site must be unique.

**Examples:**
- ✅ "Product Analytics Guide: 15 Events to Track in 2026 | ExampleCo"
- ✅ "[Tool A] vs [Tool B] vs [Tool C]: Feature + Price Comparison (2026)"
- ❌ "The Ultimate Guide to Everything You Need to Know About Product Analytics (It's Really Comprehensive!)" (too long, buries keyword, hype language)
- ❌ "Blog Post" (no keyword, no specificity)

### Meta description formula

```
[Direct value proposition] + [Specificity/proof] + [Soft CTA]
```

- **155-160 character target.** Google truncates around 920px on mobile, ~160 chars on desktop.
- **Direct value proposition** in the first sentence — what will the reader get?
- **Specificity or proof** — a number, a benefit, a differentiator.
- **Soft CTA** when appropriate — "Read the guide," "Compare tools," "Try free."
- **Do not optimize for keyword density.** Google rewrites ~60% of meta descriptions anyway; write for click-through, not for Google.

### Header hierarchy for AEO

**H1 (one per page):** Primary topic, natural phrasing, includes target keyword. Match or closely mirror the title tag.

**H2s (question-based, sequenced by user journey):** Each H2 mirrors a question a real user would ask an AI tool. LLMs treat H2s as prompts when extracting citations.

**H3s (supporting detail):** Break up long H2 sections. Not every H2 needs H3s — only when the section genuinely has sub-parts.

**H4+:** Use sparingly. If you're at H4, consider whether that section is a distinct topic that deserves its own H2.

**Direct answer pattern:** Every H2 must be followed by 1-2 sentences that directly answer the question — no throat-clearing, no hedging. This is where 78.4% of question-containing citations come from.

**Example (excellent):**
> ## How long does SEO take to show results?
>
> Most sites see initial ranking movement in 3-6 months, with meaningful traffic growth in 6-12 months. The timeline depends heavily on domain authority, competition, and technical baseline — a new site in a competitive niche often needs 12-18 months before organic traffic becomes a primary channel.

**Example (bad):**
> ## SEO Timeline
>
> SEO is a topic that many people have questions about. In this section, we'll explore the timeline and what to expect. It depends on many factors and may vary.

The first example gives the AI a clean prompt/answer pair with specific numbers. The second is unextractable.

### Body content optimization (ski ramp applied)

1. **Front-load the answer in the first paragraph.** Captures 44.2% of LLM citations.
2. **Use definitive language.** "Teams that do X see Y" beats "This may help teams understand X." LLMs penalize hedging.
3. **High entity density (~20%).** Named brands, tools, people, numbers, dates. Natural, not stuffed.
4. **Mid-paragraph information gain.** 53% of citations come from mid-paragraph, not the opener. Put the specific, concrete, citable fact in sentence 2 or 3 of a 3-5 sentence paragraph.
5. **Renders in raw HTML.** View-source test. Any critical content behind JS is invisible to AI.
6. **Summary / wake-up zone at the end.** The last 10% sees a citation bump. Include a Key Takeaways block.

### Schema markup decision matrix

| Page type | Primary schema | Additional schemas |
|-----------|---------------|---------------------|
| **Article / blog post** | `Article` (or `NewsArticle` / `BlogPosting`) | `BreadcrumbList`, `Person` (author), `Organization` (publisher), `FAQPage` (if the page has a clear FAQ) |
| **Product** | `Product` | `AggregateRating`, `Offer`, `Review`, `BreadcrumbList` |
| **How-to** | `HowTo` | `BreadcrumbList`, `VideoObject` (if video present) |
| **FAQ page** | `FAQPage` | `BreadcrumbList` |
| **Local business** | `LocalBusiness` (or specific subtype like `Restaurant`, `Plumber`) | `Service`, `GeoCoordinates`, `OpeningHoursSpecification`, `Review` |
| **Person / author** | `Person` | `sameAs` links to social/credential profiles |
| **Organization** | `Organization` | `sameAs`, `contactPoint`, `logo` |
| **Event** | `Event` | `Place`, `Offer` |
| **Recipe** | `Recipe` | `Review`, `AggregateRating`, `NutritionInformation` |
| **Video** | `VideoObject` | `Clip` (chapters) |
| **Job posting** | `JobPosting` | `Organization`, `Place` |
| **Course** | `Course` | `Organization`, `CourseInstance` |

**Implementation notes:**
- Use **JSON-LD** (not microdata or RDFa) — Google's preferred format.
- Place inside `<script type="application/ld+json">` in the `<head>` or `<body>`.
- Validate with the Rich Results Test (https://search.google.com/test/rich-results) before publish.
- **Do not mark up content that is not visible on the page.** This is a schema violation and can lead to manual action.
- Keep schema in sync with page content — stale schema (wrong price, wrong date, missing fields) hurts more than no schema.

### Internal linking rules

- **Every page links to its parent pillar.** Use the topic name as anchor where natural.
- **Every pillar links to all its clusters.** Typically in a "Deep dives" or "Related articles" section.
- **Cluster pages cross-link to 2-3 related clusters.** Contextual placement inside the body, not in a generic "related posts" widget.
- **Descriptive anchor text.** "Product analytics metrics that matter" beats "click here" or "this guide."
- **No more than ~100 unique outbound internal links per page.** Google can crawl more, but the link equity split becomes negligible.
- **Links in the main content area pass more weight** than links in sidebars or footers.

### Image optimization

- **Alt text:** Describe the image in context. Include the target keyword only if it's natural. Decorative images should have `alt=""` (empty, not missing).

From narayan-metaflow/metaflow-marketing-skills, MIT.

More from narayan-metaflow/metaflow-marketing-skills

Signals

Stars
15
Forks
2
Open issues
0
Last commit (this folder)
Apr 17, 2026 (6 months ago)

From GitHub (fetched Oct 24, 2026).

Details

License
MIT
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

More from narayan-metaflow

Done for you

Rather not set it up yourself?

Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.

See how we build AI agents →Read the case studies

Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.

launches shipped
50+
launches shipped
media placements
500+
media placements
founders on every account
2
founders on every account

Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block

Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.

Get on-page-seo-aeo-optimization running in your business

Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.

Goes to the Chalk Labs founders, not the newsletter. Privacy

Need help deploying on-page-seo-aeo-optimization?

CallGet help
Cite this page

on-page-seo-aeo-optimization skill. Agent Mart. https://agentmart.dev/skills/narayan-metaflow-on-page-seo-aeo-optimization (last updated 2026-10-24)