skills.rest
ibytechaos
ibytechaosCOMMUNITY
Explore all skills by this author →
GitHub

benchmark

Measures performance baselines and detects regressions across pages, APIs, and builds.

1|Updated Oct 11, 2025
Search Tags:#performance#benchmark#core web vitals#regression detection#ci#build performance#api latency
One-click install
npx skills add https://github.com/ibytechaos/claude --skill benchmark-ibytechaos
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: benchmark
Source: https://github.com/ibytechaos/claude/tree/main/plugins/everything-claude-code/skills/benchmark
Command: npx skills add https://github.com/ibytechaos/claude --skill benchmark-ibytechaos

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Teams often ship changes without knowing their performance impact, discovering regressions only after users complain. This Skill establishes measurable performance baselines and compares metrics before and after changes so regressions are caught before release. ## Core Features & Use Cases - Page Performance Measurement: Captures Core Web Vitals (LCP, CLS, INP, FCP, TTFB), resource sizes, and network request counts via browser MCP against defined targets. - API and Build Benchmarking: Measures endpoint latency percentiles (p50, p95, p99) under load, plus cold build, hot reload, test, lint, and type-check durations. - Before/After Comparison: Saves baselines as JSON in .ecc/benchmarks/ and produces delta tables with verdicts (better/warn) for each metric. - Use Case: Before merging a PR, run a baseline, apply your changes, then compare to see that LCP regressed by 200ms while bundle size improved by 5KB, and decide whether to ship. ## Quick Start Ask the AI to run a performance baseline of your app now, then compare it against your new changes to flag any regressions.

Frequently Asked Questions about benchmark

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

FAQPage Schema
How do I measure Core Web Vitals for my web pages?▼

Run the page performance mode, which navigates to each target URL via browser MCP and records LCP, CLS, INP, FCP, and TTFB. Results are compared against standard targets such as LCP under 2.5 seconds and CLS under 0.1.

How to detect performance regressions before merging a PR?▼

Save a baseline with the benchmark command before making changes, then run the compare mode afterward. It outputs a delta table showing each metric's before/after values with a verdict flagging regressions.

Can I benchmark API endpoint latency under load?▼

Yes, the API performance mode hits each endpoint 100 times and reports p50, p95, and p99 latency, response sizes, and status codes. It also tests with 10 concurrent requests and compares results against your SLA targets.

Where are benchmark baselines stored and shared?▼

Baselines are stored as JSON files in the .ecc/benchmarks/ directory. Because they are git-tracked, the whole team shares the same baselines and CI can run comparisons on every PR.

What are the limitations of browser-based performance measurement?▼

Browser-based metrics vary with network conditions, machine load, and caching state, so single runs can be noisy. For reliable comparisons, run benchmarks in consistent environments and pair with post-deploy monitoring.

Related Skills

Explore More Software Engineering Skills→
edercnjedercnj

x-execute-api-smoke-tests

Orchestrates black-box REST API smoke tests via Postman/Newman on deployed environments.

Community
Advanced
shogo-labsshogo-labs
12

land-and-deploy

Merge PRs, await CI, deploy, and validate production canary health.

Official
Advanced
Nairon-AINairon-AI
7

flux-epic-review

Coordinate end-to-end epic completion reviews with adversarial dual-model checks and browser QA.

Official
Advanced
Adam-GuerinAdam-Guerin
1

performance-engineer

Diagnose application bottlenecks using observability, profiling, and load testing techniques.

Community
Advanced