The agent on Vercel

How it works on Vercel

On Vercel the agent is one middleware file at the project root. It runs at the edge before the response, with any framework or none.

The file

middleware.js — At your project root. Works with or without a framework.

import { next } from '@vercel/edge'
import { reflexa } from './reflexa-agent.js'

const agent = reflexa({ token: process.env.REFLEXA_TOKEN })

export default async function middleware(request, context) {
  try {
    return (await agent.ahead(request, context)) ?? next()
  } catch {
    return next() // never let the agent take the site down
  }
}

export const config = { matcher: '/:path*' }

Get reflexa-agent.js and the site's token from the dashboard's Agent page, put the token in REFLEXA_TOKEN, deploy. Then press Check the connection in the dashboard: it fetches the domain proof from your real domain and the agent is live.

What this install serves

• counts every AI visit, page by page
• serves /llms.txt
• serves the domain proof that binds the token to this host

What it does not: it never sees the page, so robots.txt and Organization data are not switches here — the agent says so instead of pretending. Connect the site's repository in the dashboard and the agent commits those, and a page's title, description, opening and FAQ, itself.

All install guides → · what the agent never does →