benchmark

Detect performance regressions by establishing baselines for page load times and Core Web Vitals, then comparing new builds against them. The Skill applies these measurements directly to PR workflows to catch slowdowns before merging.

Track Core Web Vitals like LCP and FCP automatically across pages using the browse daemon. It captures TTFB, FCP, LCP, DOM timings, and bundle sizes to provide live performance tracking for your web application.

Baseline-first performance benchmarking establishes a historical performance reference before measuring new builds. You need it when you want to validate PRs, run Canary analyses, and track performance trend data over time to prevent slowdowns.

Generate a performance trend report by running a benchmark on a URL to establish a baseline, then comparing it against historical runs. It produces a structured local report containing baselines, diffs, and trend data for later review.

This benchmarking approach requires no external dependencies or components to run. It uses the browse daemon to measure page load times, resource sizes, and Core Web Vitals, storing the resulting structured reports locally.

Compare bundle sizes and DOM timings before and after a code change by establishing baselines and comparing new builds. It captures these metrics across pages and produces a structured report with diffs to highlight regressions.