Missing error monitoring in production
Error monitoring captures unhandled exceptions, failed requests, and slow performance in production and routes them to a dashboard you can actually look at. Without it, the first time you hear about a bug is from a user, and only if they bother to email.
#What goes wrong
Most AI scaffolds skip the monitoring step. The app ships, users hit edge cases, the server returns 500s, and no one knows. Console logs disappear into the void of serverless function output.
#Why it matters
A single broken release can hurt conversion for days before anyone notices. With monitoring, you get a Slack ping the moment the rate spikes. Without it, you ship blind. Setup takes less than fifteen minutes and pays back the first time it catches a real bug.
#How Heimdall checks for this
Heimdall reads your project files looking for Sentry, Bugsnag, Highlight, or equivalent initialization calls. It checks both client-side and server-side configuration to make sure errors from both sides are captured.
#How to fix it
Add Sentry through the official Next.js wizard, which sets up client, server, and edge configs in one go. Verify it works by throwing a test error in development. Configure source maps so stack traces are useful in production.
Frequently asked questions
Sentry has a free tier, right?
Do I need both client and server monitoring?
What about logging?
Run this check on your own repo
Heimdall scans your GitHub repo for this and 16 other issues in under a minute.