benchmark

Web performance regression detection measures current page load timings against a historical baseline to identify deterioration. This skill captures per-page metrics like TTFB, FCP, and LCP across PR workflows, generating JSON and Markdown diff reports to highlight performance drift.

You need to track TTFB, FCP, LCP, DOM Interactive, DOM Complete, and Full Load timings for frontend regression testing. Capturing these per-page metrics alongside resource and bundle analyses establishes a solid baseline to identify performance regressions.

You establish a performance baseline for web page audits by running an automated benchmark on a target URL using a browse daemon. This captures initial page load and resource metrics, creating a reference point for historical trend analysis to spot performance deterioration over time.

Yes, you can analyze bundle size and resource loading as part of PR performance checks. The benchmarking process includes resource and bundle analysis alongside page load timings, providing a comprehensive diff view to identify regressions introduced by the pull request.

Yes, benchmarking page load times works with automated historical trend analysis to monitor performance drift. By comparing current metrics like DOM Complete and Full Load against stored baselines, the system automatically identifies gradual deterioration across multiple historical audits.

The best way to generate performance regression reports for code reviews is to output structured JSON and Markdown files from your benchmarks. These reports provide diff views of timing metrics and bundle changes, making it easy to evaluate web page performance impacts directly in the PR workflow.