benchmark

Benchmark web page performance against a stored baseline using browse daemon metrics.

Updated Apr 26, 2026
One-click install
npx skills add https://github.com/XinheLIU/human-leverage --skill benchmark-xinheliu
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: benchmark
Source: https://github.com/XinheLIU/human-leverage/tree/main/.claude/skills/benchmark
Command: npx skills add https://github.com/XinheLIU/human-leverage --skill benchmark-xinheliu

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Benchmarking and regression detection for web performance using a dedicated browse daemon, enabling teams to baseline and continuously monitor page timings and resource behavior across PRs.

Core Features & Use Cases

  • Baseline capture: Save a baseline of page performance and compare it against new runs.
  • Comprehensive metrics: TTFB, FCP, LCP, DOM Interactive/Complete, and resource load details.
  • Reports and trends: Generate baselines, comparisons, and trend summaries for PR reviews and budgets.

Quick Start

Run /benchmark on a target URL to generate a baseline and begin comparison of performance metrics.

Frequently Asked Questions about benchmark

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

FAQPage Schema
How do I detect web performance regressions across multiple pages?

Detect web performance regressions by collecting navigation timing data and resource metrics via a browse daemon, then comparing the results against a stored baseline JSON file to identify timing and resource deviations.

What web performance metrics are captured during a baseline benchmark?

A baseline benchmark captures web performance metrics including TTFB, FCP, LCP, DOM Interactive, DOM Complete, and detailed resource load timings to validate page behavior and establish budgets.

How do I set up a performance baseline for PR validation?

Set up a performance baseline for PR validation by running the benchmark command on a target URL. This generates a baseline JSON file in the .gstack/benchmark-reports/baselines directory to enable future comparisons.

Do I need a browse daemon running to collect page timing data?

Yes, a running browse daemon is required to collect page timing data. You also need permission to write to the .gstack/benchmark-reports directory to save baselines, comparisons, and trend summaries.

Why does my performance comparison fail without a baseline?

Performance comparison fails without a baseline because the tool requires a baseline JSON file at .gstack/benchmark-reports/baselines/baseline.json to calculate timing differences and resource usage regressions.

Can I track resource usage and load timings alongside navigation metrics?

Yes, you can track resource usage and load timings alongside navigation metrics. The benchmark generates detailed reports combining TTFB, FCP, LCP, and resource load details for comprehensive web performance monitoring.