benchmark

Benchmark web page performance and flag regressions against baselines.

2|Updated Mar 31, 2026
One-click install
npx skills add https://github.com/jayzalowitz/skytwin --skill benchmark-jayzalowitz
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: benchmark
Source: https://github.com/jayzalowitz/skytwin/tree/main/.claude/skills/gstack/benchmark
Command: npx skills add https://github.com/jayzalowitz/skytwin --skill benchmark-jayzalowitz

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill detects performance regressions in web pages by establishing baselines for load times, Core Web Vitals, and resource usage, and by comparing new builds to those baselines.

Core Features & Use Cases

  • Baseline capture for a URL or page suite, enabling quick comparisons across builds.
  • Automatic detection of regressions in key metrics like TTFB, FCP, LCP, and total resource load.
  • Trend analysis across multiple runs to reveal slowdowns over time and inform optimization priorities.

Quick Start

Run /benchmark on a URL to start a full performance audit and baseline comparison.

Frequently Asked Questions about benchmark

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I detect performance regressions in web pages during PR workflows?

Detect web page performance regressions by capturing load time and Core Web Vitals baselines, then comparing new builds against them within PR workflows to flag slowdowns automatically.

How do I establish a performance baseline for a URL or page suite?

Establish a performance baseline by running a benchmark on a URL or page suite to capture TTFB, FCP, LCP, and total resource load metrics, which can then be saved in JSON and markdown formats.

Can I track Core Web Vitals trend analysis across multiple builds?

Track Core Web Vitals trend analysis across multiple builds by comparing per-page metrics over time, revealing slowdowns to inform optimization priorities.

Do I need the browse daemon to measure LCP and FCP metrics?

Yes, measuring LCP, FCP, and other web performance metrics requires the browse daemon to run the benchmark audits and capture page load data.

What is the best way to compare web performance against a PR baseline?

The best way to compare web performance against a PR baseline is to automatically detect regressions in key metrics like TTFB and LCP by evaluating new page loads against previously captured JSON or markdown baselines.

Why does my web performance benchmark show different results across runs?

Web performance benchmark results vary across runs due to fluctuating network conditions and resource usage, making trend analysis across multiple builds necessary to reveal true slowdowns over time.