benchmark

Measure and compare web page, API, and build performance metrics against JSON baselines.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/Maelwalser/claude-config --skill benchmark-maelwalser
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: benchmark
Source: https://github.com/Maelwalser/claude-config/tree/main/skills/benchmark
Command: npx skills add https://github.com/Maelwalser/claude-config --skill benchmark-maelwalser

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Many changes introduce unnoticed performance regressions across pages, APIs, and build processes; this Skill provides repeatable baselining and deterministic comparisons to detect and quantify regressions before they reach users.

Core Features & Use Cases

  • Page Performance: Capture Core Web Vitals (LCP, CLS, INP, FCP, TTFB), measure resource sizes, count network requests, and detect render-blocking assets for target URLs.
  • API Performance: Hit endpoints repeatedly to collect p50/p95/p99 latencies, response sizes, status code distributions, and run short concurrent-load checks.
  • Build Performance: Record cold and incremental build times, hot reload/HMR latency, test suite duration, TypeScript check time, linting, and Docker build timings.
  • Before/After Comparison & CI Integration: Save baselines, compare before/after runs, store JSON baselines in the repository, and run comparisons in CI to surface regressions on PRs.
  • Use Case: Capture a baseline before a feature branch, run the benchmark again in CI on the PR, and compare deltas to decide whether to accept the change.

Quick Start

Run the benchmark baseline command to capture current page, API, and build metrics for later 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 my CI pipeline before shipping?

Detect performance regressions in CI by capturing baseline metrics like Core Web Vitals and API latencies, storing them as JSON in your repo, and comparing before/after runs on pull requests to surface quantified deltas.

What is the best way to benchmark Core Web Vitals like LCP and CLS for web pages?

Benchmarking Core Web Vitals involves measuring LCP, CLS, INP, FCP, and TTFB for target URLs to capture resource sizes, count network requests, and identify render-blocking assets for repeatable performance auditing.

How do I measure API latency under load and capture p95 or p99 response times?

Measure API latency under load by hitting endpoints repeatedly to collect p50, p95, and p99 response times, status code distributions, response sizes, and running short concurrent-load checks to quantify performance.

Can I track build performance regressions like incremental build times and hot reload latency?

Track build performance regressions by recording cold and incremental build times, hot reload and HMR latency, test suite duration, TypeScript check times, linting, and Docker build timings to monitor developer feedback loops.

How do I compare before and after performance metrics for a feature branch?

Compare before and after performance metrics by capturing a baseline on your main branch, running the benchmark again on your feature branch in CI, and comparing the stored JSON deltas to decide whether to accept the change.

Do I need external dependencies to measure web vitals and API latencies locally?

You do not need external dependencies to measure web vitals and API latencies locally; the Skill operates independently to collect metrics, store baselines as JSON files, and execute deterministic before/after comparisons.