skills.rest
tgmarinho
tgmarinhoCOMMUNITY
Explore all skills by this author →
Website𝕏@tgmarinhoGitHub

benchmark

Detects web performance regressions by comparing page metrics against captured baselines.

Updated May 15, 2026
Search Tags:#performance#web vitals#bundle size#benchmark#regression detection#lcp#page speed
One-click install
npx skills add https://github.com/tgmarinho/canetaco --skill benchmark-tgmarinho
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: benchmark
Source: https://github.com/tgmarinho/canetaco/tree/main/.claude/skills/benchmark
Command: npx skills add https://github.com/tgmarinho/canetaco --skill benchmark-tgmarinho

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Web apps get slow gradually as each PR adds small delays and bundle weight, and nobody notices until load times are unacceptable. This Skill measures real page performance, stores baselines, and flags regressions before they ship. ## Core Features & Use Cases - Baseline Capture and Comparison: Record TTFB, FCP, LCP, DOM timings, transfer sizes, and request counts per page, then diff current metrics against the stored baseline with regression thresholds. - Resource and Bundle Analysis: List the slowest resources, JavaScript and CSS bundle sizes, and network summaries via the browse daemon's perf command and JavaScript evaluation. - Budget and Trend Checks: Grade pages against industry performance budgets and track metric trends across historical benchmark runs. - Use Case: Before merging a PR, run a diff-scoped benchmark on affected pages to catch an LCP jump from 800ms to 1600ms caused by a new unoptimized image. ## Quick Start Ask the assistant to run /benchmark on your local dev URL with the --baseline flag first, then rerun it after your changes to see a regression report.

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 before merging a PR?▼

Capture a baseline with /benchmark <url> --baseline on the main branch, then rerun /benchmark on your branch. The skill compares TTFB, FCP, LCP, bundle sizes, and request counts, flagging regressions over 50% or 500ms.

How to measure Core Web Vitals like LCP and FCP from the command line?▼

The skill uses the browse daemon's perf command plus JavaScript evaluation of the Performance API to read navigation, paint, and resource entries. It extracts FCP, LCP, TTFB, DOM Interactive, and full load times per page.

Can I benchmark only the pages changed in my branch?▼

Yes. The --diff flag runs git diff against the PR base branch to identify changed files and benchmarks only the affected pages, instead of auditing the whole site.

What thresholds count as a performance regression?▼

Timing metrics regress at over 50% increase or 500ms absolute increase, with warnings above 20%. Bundle sizes regress above 25% growth, and request counts warn above 30% growth.

Does the benchmark skill track performance trends over time?▼

Yes. The --trend mode loads historical baseline files from .gstack/benchmark-reports and shows how FCP, LCP, bundle size, and request counts evolved across recent benchmark runs.

Related Skills

Explore More Software Engineering Skills→
ClaudfatherClaudfather

qa

Automates headless-browser QA testing, visual crawls, and performance benchmarking against running web applications.

Official
Advanced
quanngynxquanngynx

vite-bundle-optimization

Optimize Vite build outputs with bundle splitting and dynamic imports.

Community
Advanced
Jia-Hong-PengJia-Hong-Peng

canary

Monitor live production apps for console errors and performance regressions after deployment.

Community
Advanced
AgentComputerAIAgentComputerAI
5

camoufox

Scrape websites headlessly with a fingerprint-spoofing Firefox fork via Playwright-core.

Official
Advanced