Diagnostics

Every live deploy is audited. A headless Chrome runs Lighthouse against the release and scores it on performance, accessibility, best practices, and SEO, alongside runtime findings the audit turned up: console errors and failed network requests. get returns that overview, the four scores, the findings, a distilled list of the failing audits, and a fresh signed URL to the full report, defaulting to the live release or scoped to any past one. report fetches the raw Lighthouse JSON behind it for a complete drill-down.

The audit runs asynchronously a short while after a release goes live, so a call right after a deploy returns a pending status until it lands; get --wait blocks until it does. This is how an app tracks its own crawler-facing performance and SEO health over time, whether that happens from the terminal or in code.