メニュー
Full SEO and AEO (Answer Engine Optimization) audit that reads your codebase, scores every signal, and fixes what's missing — meta tags, structured data, sitemap, robots.txt, llms.txt, and AI crawler rules. One command to make any web project discoverable by Google and AI search.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
Wire PostHog analytics, Sentry error tracking, and a health endpoint into any web project. Reads your stack, installs the right packages, and instruments your app so you stop flying blind in production.
Full Answer Engine Optimization audit — make your project discoverable by AI search engines (ChatGPT, Perplexity, Claude, Google AI Overviews), not just Google. Use when the user wants to improve their site's visibility in AI-powered search, add structured data, create llms.txt, optimize for AI citations, or audit AEO readiness. Triggers on requests like "AEO audit", "AI search optimization", "make my site visible to AI", "structured data", "schema markup", "llms.txt", "answer engine", "AI discoverability", or any mention of being found by AI search engines or chatbots.
Audit or generate all UI microcopy for an app — empty states, error messages, loading states, onboarding flows, tooltips, button labels, confirmation dialogs, and success messages. Use when the user wants to write UI copy, fix empty states, improve error messages, audit UX writing, make their app feel polished, or replace placeholder text. Triggers on requests like "write app copy", "fix my empty states", "error messages", "UI copy", "microcopy", "UX writing", "my app feels robotic", "placeholder text", "button labels", "onboarding copy", "toast messages", or any mention of in-app text, UI writing, or product copy.
Generate a weekly changelog card for Twitter/X. Reads git log, distills the work into 6 punchy items, fills an HTML template, and opens it in the browser ready to screenshot.
Write cold emails that actually get responses. Covers investors, employers, mentors, partners — anyone you don't know yet. Strict 2-4 sentence formula with real examples that got replies from Evan Spiegel, Mark Cuban, and Elon Musk. Triggers on requests like "cold email", "write an email to...", "reach out to...", "email this investor", "email this founder", "outreach email", "networking email", or any request to contact someone you don't have a relationship with yet.
Generate complete multi-touch cold outreach sequences — not just one email, but a full 3-5 email sequence with timing, escalation logic, and a different angle per touch. Use when the user needs a sales sequence, drip campaign, follow-up series, or any cold outreach with more than one touchpoint. Triggers on requests like "outreach sequence", "email sequence", "follow-up series", "drip campaign", "sales cadence", "multi-touch outreach", or when someone needs more than a single cold email.
| name | seo-ready |
| description | Full SEO and AEO (Answer Engine Optimization) audit that reads your codebase, scores every signal, and fixes what's missing — meta tags, structured data, sitemap, robots.txt, llms.txt, and AI crawler rules. One command to make any web project discoverable by Google and AI search. |
| category | seo |
| tags | ["seo","aeo","meta-tags","structured-data","sitemap","robots-txt","llms-txt","open-graph","schema","ai-search"] |
| author | tushaarmehtaa |
Full SEO and AEO audit that reads the codebase, scores every signal, and applies the fixes. Not just a report — it changes the files.
Identify the framework and routing pattern before doing anything else.
Check for these in order:
next.config.* → Next.js (check version in package.json — v13+ uses App Router with generateMetadata, older uses _document and next/head)astro.config.* → Astro (uses frontmatter + <head> in layouts)nuxt.config.* → Nuxt (uses useHead or useSeoMeta)svelte.config.* → SvelteKit (uses <svelte:head>)index.html at root → Static HTML / Vite SPAgatsby-config.* → Gatsby (uses gatsby-plugin-react-helmet)Find all pages/routes:
app/**/page.tsx or page.jsxpages/**/*.tsxsrc/pages/**/*.astro.html filesTell the user: "Found [framework] project with [N] pages. Running full SEO/AEO audit."
Check each of these across all pages. Score each as PASS / WARN / FAIL.
<title> or equivalent metadatametadata.title or generateMetadata() return value%s | Brand Name) in root layoutmetadata.description<h1> per page (PASS), zero or multiple (FAIL)og:title, og:description, og:type, og:url, og:image — all must be presentog:image is critical — without it, social shares show a blank cardmetadata.openGraph objecttwitter:card, twitter:title, twitter:descriptiontwitter:card should be summary_large_image for best visibilitymetadata.alternates.canonical<link rel="canonical"><script type="application/ld+json"> blocksFAQPage — captures long-tail questions in AI answers (HIGH IMPACT)HowTo — wins "how to" featured snippetsArticle / BlogPosting — for content pagesWebApplication — for SaaS productsProduct + Review — for e-commerceOrganization — for brand identityBreadcrumbList — for navigation context<script> tags or in metadatasitemap.xml or sitemap.ts (Next.js) or sitemap-index.xmllastmod, changefreq, priority/blog/[slug] pages)robots.txt or robots.ts (Next.js)/api/, /admin/, /auth/, /dashboard/)robots.txt, check for explicit rules for AI crawlers:
GPTBot (OpenAI / ChatGPT)ChatGPT-User (ChatGPT browsing)Claude-Web (Anthropic)PerplexityBot (Perplexity)Applebot-Extended (Apple Intelligence)Google-Extended (Gemini)llms.txt in the public directory<img> tags and Next.js <Image> components for alt textnext/image usage (not raw <img>) in Next.js projects"use client" on pages that could be server-rendered)Calculate the overall score:
Display the report:
SEO/AEO Readiness: [SCORE]/100 ([VERDICT])
Framework: [detected framework]
Pages scanned: [N]
On-site checks:
[PASS] Title tags All pages have titles (30-60 chars)
[WARN] Meta descriptions 2/5 pages missing descriptions
[FAIL] Structured data No JSON-LD schema detected
[PASS] Sitemap sitemap.ts found, covers all pages
...
AEO checks:
[FAIL] AI crawler rules No AI bot rules in robots.txt
[FAIL] llms.txt Not found
[WARN] Question headings 1 question heading found (need 2+)
[FAIL] FAQ schema No FAQPage schema
...
Top issues to fix (by impact):
1. (HIGH) Add FAQPage JSON-LD schema — captures AI answer citations
2. (HIGH) Create llms.txt — makes your site readable by AI tools
3. (HIGH) Add AI crawler allow rules — GPTBot, Claude-Web, PerplexityBot
4. (MEDIUM) Add meta descriptions to 2 pages
5. (LOW) Add alt text to 3 images
Then ask: "Want me to fix all [N] issues, or pick specific ones?"
Apply fixes based on the detected framework. Ask before making changes.
Next.js App Router:
metadata export in page files or generateMetadata() for dynamic pageslayout.tsx: title: { template: '%s | Brand', default: 'Brand — tagline' }Static HTML:
<meta> tags in <head> sectionNext.js:
openGraph: {
title: 'Page Title',
description: 'Page description',
type: 'website', // or 'article' for content pages
url: 'https://yoursite.com/page',
siteName: 'Brand Name',
locale: 'en_US',
},
twitter: {
card: 'summary_large_image',
title: 'Page Title',
description: 'Page description',
}
Next.js App Router:
Create opengraph-image.tsx (or .jsx) in the app directory for auto-generated OG images:
import { ImageResponse } from 'next/og'
export const runtime = 'edge'
export const alt = 'Page Title'
export const size = { width: 1200, height: 630 }
export const contentType = 'image/png'
export default async function Image() {
return new ImageResponse(
(
<div style={{
fontSize: 64,
background: '#000',
color: '#fff',
width: '100%',
height: '100%',
display: 'flex',
alignItems: 'center',
justifyContent: 'center',
}}>
Page Title
</div>
),
{ ...size }
)
}
For static sites, remind the user to create a 1200x630 image and reference it.
Next.js App Router — create app/sitemap.ts:
import type { MetadataRoute } from 'next'
export default function sitemap(): MetadataRoute.Sitemap {
const baseUrl = 'https://yoursite.com'
return [
{ url: baseUrl, lastModified: new Date(), changeFrequency: 'daily', priority: 1 },
// Add all public pages here
]
}
Read the project's pages/routes and auto-generate all entries with appropriate priorities:
For dynamic routes, use generateStaticParams data if available to include all slugs.
Next.js App Router — create app/robots.ts:
import type { MetadataRoute } from 'next'
export default function robots(): MetadataRoute.Robots {
const baseUrl = 'https://yoursite.com'
return {
rules: [
{
userAgent: '*',
allow: '/',
disallow: ['/api/', '/admin/', '/auth/', '/dashboard/'],
},
{
userAgent: 'GPTBot',
allow: ['/', '/docs/', '/blog/'],
},
{
userAgent: 'ChatGPT-User',
allow: ['/', '/docs/', '/blog/'],
},
{
userAgent: 'Claude-Web',
allow: ['/', '/docs/', '/blog/'],
},
{
userAgent: 'PerplexityBot',
allow: ['/', '/docs/', '/blog/'],
},
{
userAgent: 'Applebot-Extended',
allow: ['/', '/docs/', '/blog/'],
},
],
sitemap: `${baseUrl}/sitemap.xml`,
}
}
Adjust the allow paths based on the project's actual public routes.
Create public/llms.txt with:
# [Product Name]
## What is [Product Name]?
[One paragraph description — what it does, who it's for]
## Key Features
- [Feature 1]
- [Feature 2]
- [Feature 3]
## Pricing
[Pricing model — free, freemium, paid, per-credit, etc.]
## Links
- Website: [URL]
- Documentation: [URL]
- GitHub: [URL if open source]
Read the project's README, landing page copy, and package.json to auto-fill this.
Add FAQPage schema where FAQ content exists:
<script type="application/ld+json">
{
"@context": "https://schema.org",
"@type": "FAQPage",
"mainEntity": [
{
"@type": "Question",
"name": "Question text here?",
"acceptedAnswer": {
"@type": "Answer",
"text": "Answer text here."
}
}
]
}
</script>
Add WebApplication schema for SaaS products:
<script type="application/ld+json">
{
"@context": "https://schema.org",
"@type": "WebApplication",
"name": "Product Name",
"description": "Description",
"url": "https://yoursite.com",
"applicationCategory": "Category",
"operatingSystem": "Web",
"offers": {
"@type": "Offer",
"price": "0",
"priceCurrency": "USD"
}
}
</script>
Add Article schema for content/blog pages:
<script type="application/ld+json">
{
"@context": "https://schema.org",
"@type": "Article",
"headline": "Article Title",
"description": "Description",
"author": { "@type": "Organization", "name": "Brand" },
"publisher": { "@type": "Organization", "name": "Brand" },
"mainEntityOfPage": { "@type": "WebPage", "@id": "https://yoursite.com/page" }
}
</script>
Next.js:
alternates: {
canonical: 'https://yoursite.com/this-page',
}
<Image> components have the alt propAfter applying fixes, re-run the audit on the modified files only. Show the before/after score:
SEO/AEO Readiness: 34/100 -> 89/100
Fixed:
+ Added meta descriptions to 3 pages
+ Created sitemap.ts with 8 pages
+ Created robots.ts with AI crawler rules
+ Created llms.txt
+ Added FAQPage schema to /guides page
+ Added OG image generator
+ Fixed heading hierarchy on 2 pages
Remaining (manual):
- Add real OG image assets (currently using generated fallback)
- Write FAQ content for /pricing page
- Add alt text to 2 product screenshots (need human description)
The user can specify a mode to add domain-specific checks:
/seo-ready --mode b2b — adds checks for:
/seo-ready --mode commerce — adds checks for:
/seo-ready --mode local — adds checks for:
/seo-ready --mode saas — adds checks for: