Audit and fix a vibe-coded app so crawlers and AI search can index every public page

Elias @isbkch

In short: This prompt helps with: audit and fix a vibe-coded app so crawlers and AI search can index every public page. Nothing to fill in: copy it as it is.

Reviews & critiquesWorks with Claude
Tune for my AI

Shared publicly by Elias (@isbkch) on X. Text is as posted. X profile · Original post · Report or remove
You are a senior technical SEO engineer. Audit and fix this codebase so every public page is fully crawlable, indexable, and ranks well on Google and AI search engines (ChatGPT, Perplexity, Claude, Gemini).

PHASE 1: AUDIT FIRST (do not change code yet)
Detect the framework, router, rendering mode (CSR / SSR / SSG), and hosting target. Then report:
1. What a crawler sees on first load of each public route (raw HTML, before JS runs)
2. Missing or duplicate <title> and meta descriptions per route
3. Missing Open Graph / Twitter card tags and OG images
4. Missing robots.txt, sitemap.xml, canonical URLs
5. Heading structure (exactly one H1 per page, logical H2/H3)
6. Images without alt text, width/height, or modern formats
7. Core Web Vitals risks: large bundles, render-blocking scripts, layout shift, unoptimized fonts
8. Broken links, missing 404 page, client-only redirects
9. Pages that should be public but sit behind auth or JS-only rendering
Output a prioritized table: issue, impact (High/Med/Low), affected files, fix. Wait for my approval before Phase 2.

PHASE 2: FIX (in priority order)
- Rendering: if public pages are client-side rendered, add SSR, SSG, or build-time prerendering using the approach native to this framework. Marketing and content pages must ship real HTML.
- Metadata: unique, keyword-relevant title (≤60 chars) and description (≤155 chars) per route, generated from a single config, not hardcoded copies.
- Social: og:title, og:description, og:image (1200×630), og:url, twitter:card=summary_large_image. Generate a default OG image if none exists.
- Indexing: robots.txt (block /api, /admin, /dashboard, auth pages), auto-generated sitemap.xml with all public routes, self-referencing canonical tags, noindex on private pages.
- Structured data: JSON-LD for Organization, WebSite, and the right type for this app (SoftwareApplication, Product, Article, FAQPage, etc.).
- Semantic HTML: <header>, <main>, <nav>, <article>, <footer>; one H1 per page; descriptive link text (no "click here").
- Performance: lazy-load below-fold images, set explicit dimensions, use WebP/AVIF, preload the main font with font-display: swap, code-split heavy routes, defer third-party scripts.
- AI search: add /llms.txt summarizing what the app does, who it's for, and links to key pages. Make sure answer-worthy content (features, pricing, FAQ) is in plain HTML text, not images or JS-only components.
- Accessibility: alt text on every meaningful image, labels on form inputs, sufficient contrast. This doubles as SEO.

RULES
- Don't change app behavior, design, or anything behind login.
- Don't invent features, prices, or claims for meta copy. Ask me if unsure.
- Keep changes minimal and framework-idiomatic. No new heavy dependencies without justification.

PHASE 3: VERIFY
Show me the raw HTML a crawler receives for the homepage and one inner page, list every file changed, and give me a checklist to confirm in Google Search Console, the Rich Results Test, and PageSpeed Insights.

Fill in the [PLACEHOLDERS] before sending. Never paste real customer data into an AI tool you don't control.

Got a prompt that works for you? Add your prompt. It takes a minute.

Sourced from a public post

Some links are affiliate or sponsored: we may earn a commission at no cost to you. Sponsor this spot →

Related prompts

Browse: Reviews & critiques · More Claude prompts

Something wrong with this prompt? Contact us or submit a better one.