Skip to main content
The Vercel integration uses a Vercel log drain to forward your site’s request logs to ChatFeatured. ChatFeatured keeps the requests made by AI crawlers and search bots and shows them in Agent Analytics. Nothing changes in your code or deployments. In ChatFeatured, open Integrations and click Vercel under Agent analytics.
Vercel offers drains only on the Pro and Enterprise plans, billed by usage. On the Hobby plan, use the Cloudflare integration instead.

​
Set it up

1

Create an API key

On the Vercel integration page, click New API Key, enter a name and click Create API Key. Copy the key from the dialog. Keys start with aa_. Creating a key needs an Owner or Admin by default.Any Agent Analytics key works for every log source, so you can reuse one from Website > Agent Analytics > Settings. A new key can take up to a minute to start working.
2

Add a drain in Vercel

  1. In the Vercel dashboard, open Team Settings > Drains and click Add Drain.
  2. Choose Logs as the data type.
  3. Name the drain (for example “Agent Analytics”) and select the projects to send.
  4. Under Additional configuration for logs, check the Static Files, Edge Functions and Functions sources and the Production environment. Don’t add sampling rules, so 100% of logs are sent.
3

Point it at ChatFeatured

Select Custom Endpoint and enter:The format must be NDJSON. With JSON, the drain passes Vercel’s test but nothing is recorded.
4

Test and create

Click Test, confirm it returns a 200 response, then click Create Drain. Test only checks the URL and API key.

​
Check that it works

A browser visit won’t show anything, because ChatFeatured keeps only requests from recognized bots. Request a page with a crawler user agent instead:
Vercel batches drain deliveries, so allow a minute or two. The visit then appears in Website > Agent Analytics as a GPTBot request.

​
What gets recorded

ChatFeatured reads the request log lines from the Static Files, Edge Functions and Functions sources and keeps requests from AI crawlers and search engine bots (see what gets counted). Build logs and console output are skipped. Unlike the Cloudflare Worker, the drain records every path and content type, including robots.txt, llms.txt and sitemaps. Drain Production only. Preview requests aren’t labelled by environment, so they’d mix in with production.

​
Troubleshooting

Missing API key means the header didn’t arrive. Invalid API key means the key is mistyped, deleted, or less than a minute old. Check the header is exactly X-API-Key: aa_....
Check that the drain format is NDJSON, the Static Files, Edge Functions and Functions sources are checked, and the project you tested is selected on the drain.