Introduction

Why Nuxt AI Ready?

Search is changing. Outside of search engines, people now get answers directly from ChatGPT, Claude, and other AI assistants. When these LLMs answer questions about topics related to your site, you want your content to be the source they cite to drive traffic and engagement back to you.

For that to happen, AI systems need to understand your content. New standards are being shaped like llms.txt for discoverability and MCP for letting agents query your site directly. But these standards are still evolving, and implementing them correctly can be complex and time-consuming.

• 📈 Increase citations by LLMs: AI assistants pull from sources they can parse easily. Structured, AI-friendly content gets referenced more often.
• 🔗 Direct site queries for LLMs: MCP support means assistants can pull live data from you, not just static snippets from their training.

Nuxt AI Ready converts your indexable pages into clean markdown that AI systems can consume, generates the right artifacts at build time, and serves AI-friendly formats to bots automatically.

Ready to get started? Check out the installation guide.

Features

📄 llms.txt Generation

Auto-generates the llms.txt standard files at build time:

• llms.txt — Concise site overview with key page links
• llms-full.txt — Complete markdown content for deeper context
• Content Signals — Metadata hints that tell AI systems how to use each page

🤖 AI-Friendly Content Serving

Your pages, optimized for how AI systems consume content:

• On-Demand Markdown — Any route available as .md (e.g., /about → /about.md)
• Smart Bot Detection — Serves markdown to AI crawlers automatically, HTML to everyone else
• RAG-Ready Output — Pre-chunked content for semantic search and AI chat pipelines

⚡ MCP Integration

Expose your site as a tool AI agents can query directly via the Model Context Protocol:

• Agents fetch live data instead of relying on stale training snapshots
• Define what actions and queries your site supports
• Works with Claude, ChatGPT plugins, and MCP-compatible assistants