Vercel / Next.js

One middleware. Every AI agent handled.

Add @inception-agents/vercel to your Next.js project. Detection and optimization at the edge.

View Integration Guide Get API Key

Integration Guide

Complete setup instructions

Every code block below is complete and copy-paste-ready. Total time: ~10 minutes.

Prerequisites

  • + Node.js 18+ installed
  • + Next.js 14+ project deployed on Vercel
  • + Inception Agents API key from your dashboard

Step 1: Install the SDK

npm install @inception-agents/vercel

Step 2: Set environment variable

Add to your Vercel project (via CLI or dashboard):

vercel env add INCEPTION_API_KEY
# Paste your API key when prompted

Or add to .env.local for local development:

echo "INCEPTION_API_KEY=your_api_key_here" >> .env.local

Step 3: Create the middleware

Create middleware.ts in your project root (next to package.json):

// middleware.ts
// Inception Agents — AI agent detection and content optimization
// This middleware intercepts all requests, detects AI agents, and
// serves optimized content. Human visitors pass through with <2ms overhead.

import { createInceptionMiddleware } from '@inception-agents/vercel';
import { NextResponse } from 'next/server';
import type { NextRequest } from 'next/server';

const inception = createInceptionMiddleware({
  apiKey: process.env.INCEPTION_API_KEY!,
  // Optional configuration:
  // debug: process.env.NODE_ENV === 'development',
  // excludePaths: ['/api/', '/_next/', '/static/'],
  // customDetectionRules: [],
});

export async function middleware(request: NextRequest) {
  return inception(request);
}

export const config = {
  // Match all paths except static assets and Next.js internals
  matcher: ['/((?!_next/static|_next/image|favicon.ico|robots.txt|sitemap.xml).*)'],
};

Already have a middleware.ts?

Wrap your existing logic:

// middleware.ts — with existing middleware logic
import { createInceptionMiddleware } from '@inception-agents/vercel';
import { NextResponse } from 'next/server';
import type { NextRequest } from 'next/server';

const inception = createInceptionMiddleware({
  apiKey: process.env.INCEPTION_API_KEY!,
});

export async function middleware(request: NextRequest) {
  // Inception Agents runs first — if it detects an AI agent,
  // it returns an optimized response. Otherwise, falls through.
  const inceptionResponse = await inception(request);
  if (inceptionResponse) return inceptionResponse;

  // Your existing middleware logic here
  return NextResponse.next();
}

export const config = {
  matcher: ['/((?!_next/static|_next/image|favicon.ico).*)'],
};

Step 4: Deploy

vercel --prod

Or push to your connected Git repository. The middleware activates automatically.

Step 5: Verify

After deployment, test with curl:

# Test AI agent detection (should return enriched content)
curl -s -H "User-Agent: GPTBot/1.0" \
  https://your-site.vercel.app/ | head -50

# Test human passthrough (should return normal site)
curl -s -H "User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7)" \
  https://your-site.vercel.app/ | head -50

# Test llms.txt
curl -s https://your-site.vercel.app/llms.txt

Configuration Options

Option Type Default Description
apiKey string (required) Your Inception Agents API key
debug boolean false Log detection results to console
excludePaths string[] ['/api/', '/_next/'] Paths to skip detection on
detectionThreshold number 0.70 Confidence threshold for agent detection
cacheMaxAge number 300 Cache TTL for optimized content (seconds)
enableLlmsTxt boolean true Serve /llms.txt and /llms-full.txt
enableJsonLd boolean true Inject enhanced JSON-LD for agents
enableAgentCard boolean false Serve /.well-known/agent.json

What Happens Next

  1. 1 Visit your dashboard to see AI agent traffic
  2. 2 Your site content is automatically analyzed and optimized content variants are generated
  3. 3 The learning engine begins tracking which content strategies work best for each agent type
  4. 4 Check the Optimization tab for specific recommendations to improve your AI visibility

Troubleshooting

Middleware not running

  • Ensure middleware.ts is in the project root (same level as package.json)
  • Check that the matcher config isn't excluding the paths you're testing
  • Run vercel logs to check for errors

No agent traffic showing in dashboard

  • It may take a few hours for real AI agents to visit your site
  • Use the curl commands above to simulate agent visits and verify the integration works
  • Check that your API key is correct in environment variables

Increased latency for human visitors

  • Normal overhead is <2ms. If higher, check that excludePaths covers high-traffic API routes
  • The middleware makes zero external network calls for human traffic — detection is local pattern matching

Uninstalling

npm uninstall @inception-agents/vercel
# Delete middleware.ts (or remove the inception wrapper)
vercel env rm INCEPTION_API_KEY

Framework Support

Next.js 14+

Middleware runs at the edge on Vercel

Next.js 13

Supported with App Router or Pages Router

SvelteKit 2+

Via hooks.server.ts adapter (coming soon)

Nuxt 3+

Via server middleware adapter (coming soon)

npm install and deploy.

Your Next.js site is 10 minutes away from being optimized for every AI agent.

Get Your API Key View All Integrations