Eirene

Skill markbaindesign/bain-studio/.claude/skills/eirene

Performance check only — Core Web Vitals, load time, image optimisation, caching. Returns a performance audit with Pass / Fail per metric. Use when you need a performance audit without the full Themis QA suite.From its SKILL.md

Install

npx -y skills add markbaindesign/bain-studio --skill eirene

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

2.1 KB, 527 tokens by cl100k_base, as published. Nobody here has run it

Eirene — Performance Checker

Peace arrives only when everything works as it should. Eirene measures it.

Steps

1. Identify the subject

Accept a live URL or a codebase path. If a URL is provided, run Lighthouse if available.

npx lighthouse {url} --output json --quiet 2>/dev/null | python3 -c "
import sys, json
d = json.load(sys.stdin)
cats = d['categories']
audits = d['audits']
for k in cats: print(k, round(cats[k]['score']*100))
for m in ['first-contentful-paint','largest-contentful-paint','total-blocking-time','cumulative-layout-shift','interactive']:
    a = audits.get(m, {})
    print(m, a.get('displayValue','n/a'))
"

If no URL is available, audit the code for performance anti-patterns.

2. Core Web Vitals

Metric	Target	Actual	Pass / Fail
LCP (Largest Contentful Paint)	≤ 2.5s
INP (Interaction to Next Paint)	≤ 200ms
CLS (Cumulative Layout Shift)	≤ 0.1
FCP (First Contentful Paint)	≤ 1.8s
TTFB (Time to First Byte)	≤ 800ms

3. Additional checks

Check	Pass / Flag / Fail	Note
Images served in modern format (WebP / AVIF)
Images have explicit width and height attributes
No render-blocking resources
JS bundle size reasonable for scope
Caching headers set appropriately
Mobile performance (throttled 4G simulation)

4. Verdict

PASS — all Core Web Vitals green. CONDITIONAL PASS — marginal scores, clear remediation path, no hard fails. FAIL — one or more Core Web Vitals red. List specific fixes for Hephaestus: file name, what to change, expected improvement.

Do not issue vague recommendations. "Optimise images" is not an action item. "Convert /wp-content/uploads/hero.jpg (1.2MB) to WebP — expected LCP improvement ~0.8s." is.

What ships with it

Read from the repository

Just SKILL.md. No reference files, no scripts.

Gives 0 of the 12 instructions most performance cost skills give in 527 tokens

Counted across 803 of the 1,058 authors here whose files we hold, read 2026-08-07

Keep skill files under 500 lines or tokensin 82 of 803, across 16 files
Use imperative form in instructionsin 80 of 803, across 9 files
Draft assertions while test runs are in progressin 75 of 803, across 9 files
Create two to three realistic test promptsin 74 of 803, across 9 files
Write skill descriptions to be pushyin 72 of 803, across 7 files
Save test cases to evals JSONin 72 of 803, across 6 files
Ask questions about edge cases and input formatsin 72 of 803, across 7 files
Save timing data immediately when runs completein 70 of 803, across 5 files
Include all trigger conditions in the skill descriptionin 69 of 803, across 3 files
Launch all test runs in a single turn or simultaneouslyin 69 of 803, across 3 files
Capture intent before writing a skillin 67 of 803, across 1 file
Import directly instead of barrel filesin 52 of 803, across 15 files

Said here and by no other author read

run lighthouse for a provided url
audit code for performance anti-patterns without a url
check images for modern formats
check images for explicit width and height
check for render-blocking resources
check js bundle size

Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.