Analyze browser performance data to detect regressions across pull requests.
npx skills add https://github.com/zz8011/harnesstest --skill benchmark-zz8011
Please help me install this Agent Skill. Skill: benchmark Source: https://github.com/zz8011/harnesstest/tree/main/.claude/skills/gstack/benchmark Command: npx skills add https://github.com/zz8011/harnesstest --skill benchmark-zz8011
Performance regression detection and baseline establishment for web apps, using the browse daemon to measure and compare metrics across PRs.
Run the benchmark workflow against a target URL to capture metrics and compare them to baselines.
High-intent search queries and answers about installing and using this skill.
To detect web performance regressions across pull requests, you analyze browser performance data using a benchmarking workflow that establishes baselines for Lighthouse and Web Vitals metrics. It then compares current PR measurements against these stored baselines to surface slowdowns.
When establishing a page load baseline, the tracked core web vitals include TTFB, FCP, LCP, and DOM events. The benchmark also captures resource-size breakdowns to monitor overall page performance and resource loading changes over time.
To ensure a UI change doesn't slow down page load before merging, run a benchmark workflow against your target URL. It captures current metrics and compares them to established baselines, alerting you to any page load or resource-size regressions.
Yes, this benchmarking approach works without external dependencies. It relies on a built-in browse daemon to capture browser performance data, measuring web vitals and resource sizes directly to establish baselines and detect regressions.
The best way to track resource-size changes over time for web apps is by capturing resource breakdowns through a browse daemon. This stores the data as baselines, allowing you to compare subsequent page load measurements and detect size regressions across PRs.
tropeksMeasure web performance metrics and compare them against baselines.
SoSoBurnerMeasure page load times, Core Web Vitals, and resource sizes in CI/CD pipelines.
NITRR-OfficialProfile and optimize CampusOS performance metrics, bundle size, and Lighthouse scores.
CarbeneAIRun baseline-aware browser benchmarks against URLs and report performance regressions.
