# Inlay

> Inlay makes websites discoverable, accessible, and actionable for AI agents, LLMs, and tools like Claude, ChatGPT, and Perplexity.

## What Inlay Does

Inlay helps website owners optimize their sites for AI agent discovery. Just as SEO optimizes sites for search engines, Inlay optimizes sites for AI agents — the new way people find information.

The platform provides a comprehensive suite of tools:

### AI Readiness Audit
A free 11-point audit that scores websites across:
1. **Structured Data** (15% weight) — JSON-LD schema coverage and completeness
2. **LLMs.txt** (12% weight) — AI-readable site summary file
3. **Semantic HTML** (12% weight) — Heading hierarchy, landmark elements, ARIA roles
4. **Content Accessibility** (12% weight) — Server-rendered content for AI crawlers
5. **Meta Quality** (10% weight) — Open Graph, Twitter Cards, canonical URLs
6. **AI Crawler Policy** (8% weight) — robots.txt rules for AI bots (GPTBot, ClaudeBot, etc.)
7. **API Discoverability** (8% weight) — Sitemap, RSS feeds, and API documentation
8. **Security & Trust** (8% weight) — HTTPS, security headers, status codes
9. **MCP Readiness** (5% weight) — Model Context Protocol server detection
10. **Content Freshness** (5% weight) — Date signals and update recency
11. **Image & Media** (5% weight) — Alt text, lazy loading, responsive images

Each category produces detailed findings (positive, negative, neutral) and actionable recommendations with effort estimates.

### Hosted MCP Server
A managed Model Context Protocol (MCP) server that lets AI agents like Claude programmatically access a site's content. The MCP server is hosted by Inlay — site owners don't need to run any infrastructure. Each site gets a unique MCP endpoint that serves tools for querying the site's pages, content, and metadata.

### Auto-Generated llms.txt
Automatically creates and hosts an llms.txt file — a machine-readable summary of the site following the llms.txt specification. The file is kept in sync with the site's content and includes pages, descriptions, and links. Both llms.txt (summary) and llms-full.txt (detailed) variants are served.

### Embed Script
A lightweight async script tag that, when added to a site's `<head>`, auto-injects:
- Structured data (JSON-LD) — Organization, WebSite, BreadcrumbList schemas
- AI discovery meta tags — MCP server, ai-plugin, llms-txt links
- Open Graph and Twitter Card meta tags
- Canonical URLs
- Content freshness signals (article:modified_time)

The script is non-blocking and detects existing tags to avoid duplication.

### Analytics Dashboard
Tracks how AI agents interact with a site:
- Which MCP tools agents use
- What content they access
- AI readiness score trends over time
- Engagement metrics from the embed script beacon

### Platform Plugins
One-click integration for:
- **WordPress** — Full plugin with settings page, auto-generated /llms.txt endpoint, structured data injection, MCP meta tags
- **Shopify** — Theme app extension, product/article/organization structured data via Liquid
- **Wix** — Velo (Dev Mode) integration and zero-code Custom Code snippet, SPA navigation handling
- **Squarespace** — Code Injection snippet and Developer Mode template region file

## API

### Public Endpoints (no authentication required)

**Run an AI Readiness Audit**
```
POST /api/audit
Content-Type: application/json
Body: { "url": "https://example.com" }
```
Returns: overall score (0-100), category breakdown with individual scores/weights, detailed findings, and actionable recommendations.

**Retrieve Audit by ID**
```
GET /api/audit/:id
```
Returns: previously completed audit results.

### Authenticated Endpoints (require API key)

**LLMs.txt**
```
GET /api/llms/:siteId
GET /api/llms/:siteId/full
```
Returns: llms.txt or llms-full.txt content for the given site.

**MCP Server**
```
POST /api/mcp/:siteId
```
Model Context Protocol endpoint for AI agent tool use.

**Sitemap**
```
GET /api/sitemap/:siteId
```
Returns: XML sitemap for the given site.

**Atom Feed**
```
GET /api/feed/:siteId
```
Returns: Atom feed of the site's pages.

## Pricing

| Plan | Price | Sites | Key Features |
|------|-------|-------|-------------|
| Free | $0/mo | — | 5 audits/month, detailed findings and recommendations, shareable results |
| Lite | $9/mo | 1 | Hosted MCP server, embed script, 20 audits/month |
| Starter | $29/mo | 3 | llms.txt, structured data injection, AI discovery tags, 10k requests/month |
| Pro | $99/mo | 10 | Platform plugins, analytics, AI agent insights, 100k requests/month |
| Enterprise | Custom | Unlimited | Custom MCP tools, SLA, dedicated support |

## Technical Details

- **Stack**: Next.js (App Router), TypeScript, Tailwind CSS, PostgreSQL, Drizzle ORM
- **Hosting**: Vercel
- **Domain**: inlay.dev (www.inlay.dev)
- **MCP Endpoint**: www.inlay.dev/api/mcp/:siteId
- **OpenAPI Spec**: inlay.dev/openapi.json

## Links

- Website: https://inlay.dev
- Free Audit: https://inlay.dev/audit
- Pricing: https://inlay.dev/pricing
- Contact: https://inlay.dev/contact
- OpenAPI: https://inlay.dev/openapi.json
- MCP Discovery: https://inlay.dev/.well-known/mcp.json
- LLMs.txt: https://inlay.dev/llms.txt
- LLMs Full: https://inlay.dev/llms-full.txt