GEO AI – AI Search Optimization

Universal TypeScript engine for optimizing websites for AI search engines.

A zero-dependency TypeScript engine that optimizes websites for AI search engines like ChatGPT, Claude, Gemini, Perplexity, DeepSeek, Grok, YandexGPT, GigaChat, and more. The core works with any Node.js environment — official integrations currently include Next.js and NestJS, with WordPress and Shopify available as separate ecosystem packages.

• geo-ai-core – Universal engine (llms.txt generation, bot rules, crawl tracking, caching, encryption, SEO signals, AI descriptions)
• geo-ai-next – Next.js wrapper: static file generation, middleware + route handler
• geo-ai-nest – NestJS wrapper: dynamic module, middleware, controller, guard, interceptor, decorators
• geo-ai-cli – CLI tool: geo-ai init / generate / validate / inspect for any Node.js project

Not sure what's what?

• geo-ai-core — open-source npm library, zero dependencies, works anywhere Node.js runs
• geo-ai-next — open-source npm library, Next.js integration built on top of geo-ai-core
• geo-ai-nest — open-source npm library, NestJS integration built on top of geo-ai-core
• geo-ai-cli — open-source npm CLI, geo-ai init / generate / validate / inspect for any project
• geoai.run — the analyzer, docs, and llms.txt specification site

Try the analyzer at geoai.run/analyze

Traditional SEO tools are built for search engines like Google. But AI search engines work differently — they read structured content, not meta keywords.

GEO AI provides infrastructure for AI Search Optimization:

• ✅ llms.txt generation — structured content for AI crawlers
• ✅ AI crawler management — per-bot allow/disallow rules for 16+ bots
• ✅ AI metadata — meta tags, Link headers, JSON-LD signals
• ✅ Structured signals — pricing, availability, variants in machine-readable format

 GEO AI │ ┌──────────────┼──────────────┐ │ │ │ WordPress Shopify Node.js │ │ │ GEO-AI-Woo GEO-AI-Shopify geo-ai-core │ │ │ WooCommerce geo-ai-next geo-ai-cli 

npm install geo-ai-core

import { createGeoAI } from 'geo-ai-core'; const geo = createGeoAI({ siteName: 'My Site', siteUrl: 'https://example.com', provider: { Pages: [{ title: 'Home', url: '/', description: 'Welcome' }], }, }); // Generate llms.txt const llmsTxt = await geo.generateLlms(false); // Generate llms-full.txt const llmsFullTxt = await geo.generateLlms(true);

Generates llms.txt and llms-full.txt from any data source via the ContentProvider interface. Standard version includes resource links with descriptions; full version adds content, pricing, availability, and variants.

Built-in registry of 16+ AI crawlers with per-bot allow/disallow rules and robots.txt block generation.

Pluggable cache via CacheAdapter interface with built-in implementations:

• MemoryCacheAdapter — In-memory with TTL and automatic expired entry eviction (configurable maxEntries, default 1 000)
• FileCacheAdapter — File-based with JSON metadata

GDPR-compliant bot visit logging with SHA-256 IP anonymization (Web Crypto API, Edge Runtime compatible). Pluggable storage via CrawlStore interface with built-in MemoryCrawlStore (configurable maxEntries, default 10 000).

Optional module (geo-ai-core/ai) for generating AI-optimized descriptions via Claude or OpenAI APIs:

• Customizable prompt templates with {title}, {content}, {type}, {price}, {category} placeholders
• Bulk generation (up to 50 items, batched by 5 concurrently) with progress callback
• Sliding window rate limiter (default 10 req/min)
• Typed error classification (auth, rate limit, server, network)

• <meta name="llms"> and <meta name="llms-full"> tags
• HTTP Link header for AI content discovery
• JSON-LD Schema.org (WebSite, Product, Article)

AES-256-GCM encryption for API keys via node:crypto. Format: base64(IV[12] + authTag[16] + ciphertext).

npm install geo-ai-core # or for Next.js projects: npm install geo-ai-next # or for NestJS projects: npm install geo-ai-nest # or CLI (any project): npm install --save-dev geo-ai-cli # or globally: npm install -g geo-ai-cli

import { createGeoAI } from 'geo-ai-core'; const geo = createGeoAI({ siteName: 'My Site', siteUrl: 'https://example.com', provider: { Products: [ { title: 'Widget', url: '/products/widget', description: 'A great widget' }, ], Blog: [ { title: 'Hello World', url: '/blog/hello', description: 'First post' }, ], }, }); // Generate llms.txt const llmsTxt = await geo.generateLlms(false); // Generate llms-full.txt const llmsFullTxt = await geo.generateLlms(true); // Generate robots.txt block const robotsTxt = geo.generateRobotsTxt(); // SEO signals const metaTags = geo.generateMetaTags(); const linkHeader = geo.generateLinkHeader(); const jsonLd = geo.generateJsonLd();

import { createGeoAI, type ContentProvider } from 'geo-ai-core'; class StrapiProvider implements ContentProvider { async getSections(options?: { locale?: string }) { const products = await fetchProducts(options?.locale); return [ { name: 'Products', type: 'product', resources: products }, { name: 'Blog', type: 'page', resources: await fetchPosts() }, ]; } } const geo = createGeoAI({ siteName: 'My Site', siteUrl: 'https://example.com', provider: new StrapiProvider(), crawlTracking: true, });

Generate public/llms.txt and public/llms-full.txt before next build:

// scripts/generate-llms.ts import { generateLlmsFiles } from 'geo-ai-next'; await generateLlmsFiles({ siteName: 'My Site', siteUrl: 'https://example.com', provider: new MyProvider(), });

{ "scripts": { "geo:generate": "geo-ai-generate", "build": "npm run geo:generate && next build" } }

Next.js serves files from public/ automatically — no middleware needed. Middleware and route handler still work for dynamic use cases.

// middleware.ts import { geoAIMiddleware } from 'geo-ai-next'; export default geoAIMiddleware({ siteName: 'My Site', siteUrl: 'https://example.com', provider: new MyProvider(), cache: '24h', }); export const config = { matcher: ['/((?!_next/static|_next/image|favicon.ico).*)'], };

// app/llms/route.ts import { createLlmsHandler } from 'geo-ai-next'; export const { GET } = createLlmsHandler({ siteName: 'My Site', siteUrl: 'https://example.com', provider: new MyProvider(), cacheMaxAge: 3600, // seconds, default 3600 });

Register GeoAIModule once at the application root — middleware, controller, guard, interceptor, and decorators are all available automatically:

// app.module.ts import { GeoAIModule } from 'geo-ai-nest'; @Module({ imports: [ GeoAIModule.forRoot({ siteName: 'My Site', siteUrl: 'https://example.com', provider: new MyProvider(), isGlobal: true, injectLinkHeader: true, // inject Link header on all responses }), ], }) export class AppModule {}

Async configuration with useFactory:

GeoAIModule.forRootAsync({ imports: [ConfigModule], inject: [ConfigService], useFactory: (config: ConfigService) => ({ siteName: config.get('SITE_NAME'), siteUrl: config.get('SITE_URL'), provider: new MyProvider(), cache: '24h', }), })

Use GeoAIService anywhere via DI, protect routes with GeoAIGuard, and inject the detected bot name with @IsAIBot():

import { GeoAIService, GeoAIGuard, IsAIBot } from 'geo-ai-nest'; @Controller('ai') @UseGuards(GeoAIGuard) // allow only verified AI bots export class AiController { constructor(private readonly geoAI: GeoAIService) {} @Get('meta') getMeta(@IsAIBot() bot: string | null) { return { bot, meta: this.geoAI.generateMetaTags() }; } }

/llms.txt, /llms-full.txt, and /.well-known/llms.txt are served automatically by the built-in middleware. /robots-ai.txt is served by the built-in controller.

Generate and validate llms.txt without writing any code:

# Scaffold config npx geo-ai init # Generate llms.txt + llms-full.txt into ./public npx geo-ai generate # Validate local files npx geo-ai validate # Validate remote deployment npx geo-ai validate --url https://example.com # Preview config before generating npx geo-ai inspect

See geo-ai-cli for full documentation.

import { AiGenerator } from 'geo-ai-core/ai'; const ai = new AiGenerator({ provider: 'anthropic', apiKey: 'sk-...', model: 'claude-sonnet-4-20250514', }); const description = await ai.generate({ title: 'Premium Widget', content: 'A high-quality widget made from...', type: 'product', price: '$29.99', });

interface GeoAIConfig { // Required siteName: string; siteUrl: string; provider: ContentProvider | Record<string, Resource[]>; // Optional siteDescription?: string; crawlers?: Record<string, 'allow' | 'disallow'> | 'all'; cache?: CacheAdapter | string; // '1h', '24h', '7d' or custom adapter crypto?: { encryptionKey: string }; // 64-char hex crawlTracking?: { store?: CrawlStore; secret?: string } | true; }

• Node.js 20 or higher
• TypeScript 5.5+ (recommended, ships .d.ts)

# Clone git clone https://github.com/madeburo/GEO-AI.git cd GEO-AI/geo-ai-core # Install npm install # Build all packages npm run build # Run tests npx vitest run # Type check npx tsc --noEmit

Contributions are welcome! Please see CONTRIBUTING.md for guidelines.

GEO AI Core is open-source software licensed under the MIT.

• Author: Made Büro
• Website: geoai.run
• GitHub: @madeburo
• X: @imadeburo