Live demo
Search This Site the Way a RAG Agent Would
Ask a real question. This calls /api/ai-search — the live retrieval endpoint (Cloudflare AI Search) indexed over this site's own pages — and returns the actual pages it's grounded in, not a generated answer with no source.
What happens outside the chat window
Why this matters
An AI agent that answers from general training data will confidently make things up about your business — pricing that changed, a policy that no longer applies. Grounding every answer in your actual, current content is the difference between a chatbot that sounds right and one that is right. See what AI actually cites (real data on markdown vs. HTML) for why this site's content is structured the way it is, or the internal knowledge agent pattern for the client-facing version of this same grounding.
What just happened
Your question was sent as a real vector search over every page on this site (Cloudflare AI Search), and the results above are the actual source pages it matched — the same grounding a RAG agent would use to answer instead of guessing from general training data. See RAG development for what building this for your own documents looks like.
Query it directly
GET /api/ai-search?q=<question> — no API key, JSON response. Full details on the API reference page.
The other live demo
Try the MCP playground — call real tools on the site's live MCP server instead of searching its content.
Have a product in mind? Let's scope it this week.
A 30-minute call gets you a rough timeline and cost — no obligation, no sales deck.