Lighthouse, with your code attached.
Performance auditing, Real User Monitoring, and actionable recommendations β all inside your Laravel app. Your data stays in your own database. No SaaS, no per-seat billing.
Everything Lighthouse can tell you, plus what your server was doing at the same moment.
Most performance tools stop at the page load. Laravel Vitals goes one step further β it correlates each Lighthouse finding with the queries, memory, and code paths that produced it.
Lighthouse-grade audits, but they also know what your server was doing.
Most performance tools stop at the page load. Pick what to compare us against β the gap is the same shape every time.
"Eliminate render-blocking resources" β fine, but which line?
Seven analyzers scan your project: Blade asset tags, image tags, Laravel config settings, Composer packages, Vite config, Blade view patterns, and .env settings β attaching the exact file and line to each Lighthouse finding.
- Blade asset tags β render-blocking css/js detection
- Image tags β defer offscreen images, image-alt missing
- Laravel config β missing config:cache, debug-in-prod, sync queue, disabled OPcache
- Composer packages β outdated package-level recommendations
- Vite config β unused/uncompressed bundle hints
- Blade view patterns β N+1 caller file + line, top 3 patterns
- .env settings β debug / config / locale checks
Your data never leaves your database.
Audits, telemetry, RUM beacons β all stored in your own Laravel app. No IP addresses, no cookies, no session identifiers. No SaaS account. No per-seat billing. MIT licensed, drop-in, gone if you uninstall.
Click through a fully seeded dashboard before you install.
Every page is reachable β Overview, URLs, Issues, SEO, RUM, Queries, Learn, Budgets. Real data shape, fake hostnames. Drop in, click around, share the link with your team.
- 8 dashboard pages
- Light + dark mode
- Spotlight βK
- No login required
Get to the bottom of every Lighthouse score.
Install the package, declare your URLs, run an audit. You'll have backend telemetry and code-level recommendations in under five minutes.