Skip to content
Agentmart
Submit a tool

perf-budget

Check a static brochure site against the static-web performance budget: Lighthouse mobile 90+, page weight under 1MB, WebP with srcset, self-hosted fonts with…

Skill by jeanchastel

IndexedMIT
Write a review

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/jeanchastel-perf-budget

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

Check a static brochure site against the static-web performance budget: Lighthouse mobile 90+, page weight under 1MB, WebP with srcset, self-hosted fonts with font-display swap, and no render-blocking third-party scripts. Use local or live mode before launch, after assets, or for a regression check.

Use cases: finance

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.

Tools
No tools found
Secrets
None asked for
Install scripts
None found

Network hosts it talks to 1

  • codeload.github.com · found in install command

Files it touches

  • ~/.claude/skills/perf-budget

What's inside

Skill name
perf-budget
Description
Check a static brochure site against the static-web performance budget: Lighthouse mobile 90+, page weight under 1MB, WebP with srcset, self-hosted fonts with font-display swap, and no render-blocking third-party scripts. Use local or live mode before launch, after assets, or for a regression check.
License
MIT
Instructions
76 lines in SKILL.md · view on GitHub

Files (1)

  • SKILL.md

SKILL.md (first lines)

# Performance Budget Check

Enforce the static-web performance budget. Two modes: local static analysis of the repo, or live Lighthouse against a deployed URL.

## Profile gate

Use for repos with `profile: static-web`; see `static-web/static-web-standard.md` for the profile gate.

## When to Use

- Before launch.
- After dropping in new hero imagery or third-party scripts.
- Monthly as a regression check on the live site.
- When the user says "check performance", "is this fast", "audit Lighthouse".

For UI breakage at breakpoints, use responsive. For markup quality, use html-audit. This skill cares about weight, format, and what the browser does on first paint.

## How It Works

When a deploy-prep shared working set is present, evaluate the performance lens against that set and do not re-enumerate or re-read files. Standalone invocation still self-scans `**/*.html`, `**/*.css`, `**/*.js`, and `assets/**`.

### Budget

- Lighthouse mobile performance: 90 or higher. Below blocks ship.
- First contentful paint under 2 seconds on simulated 4G.
- Home page total weight under 1MB, excluding optional below-fold hero video.
- All raster images: WebP, with `srcset`, lazy-loaded below the fold.
- All fonts: self-hosted, `font-display: swap`, subset where practical.
- No render-blocking third-party scripts in `<head>`. Analytics is the typical offender; load it `defer` or `async`.

### Local mode

Walk the repo. Per page (each `.html`):

1. Sum the weight of every asset referenced directly or transitively (images, fonts, CSS, JS).
2. Flag raster images larger than 200KB or in PNG/JPG when WebP would serve.
3. Flag `<img>` without `srcset` or without `loading="lazy"` where below the fold.
4. Flag Google Fonts and other external font CDNs.
5. Flag external scripts in `<head>` without `defer` or `async`.
6. Flag inline `<style>` or `<script>` blocks larger than 4KB (they should probably be external).

### Live mode

If given a URL, run Lighthouse mobile and report Core Web Vitals (LCP, INP, CLS) against the budget. If Lighthouse is unavailable, fall back to local mode and note the missing live data.

### Output

Local:

```
Page: index.html
  Weight:        842 KB              PASS  (<1MB)
  Images:        7 (3 WebP, 4 PNG)   FAIL  - convert PNG to WebP
  Lazy load:     5/7 below-fold      FAIL  - add loading="lazy" to 2
  Fonts:         self-hosted, swap   PASS
  Render block:  1 external script   FAIL  - add defer
```

Live:

```
Lighthouse mobile: 88 / 100  FAIL (target 90)
  LCP: 2.4s  budget: 2.5s  PASS
  CLS: 0.12  budget: 0.10  FAIL
  INP: 180ms budget: 200ms PASS
```

End with: `Budget met` or `Budget missed: <count> items`.

## Cross-references

- `static-web/static-web-standard.md`, section "Performance budget".
- responsive for srcset behavior across breakpoints.
- deploy-prep, which aggregates this verdict into the launch gate.

From jeanchastel/arkira, MIT.

More from jeanchastel/arkira

Signals

Stars
0
Forks
0
Open issues
0
Last commit (this folder)
Oct 5, 2026 (24 days ago)
Last release
Oct 23, 2026 (6 days ago)

From GitHub (fetched Oct 24, 2026).

Details

License
MIT
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

More from jeanchastel

Cite this page

perf-budget skill. Agent Mart. https://agentmart.dev/skills/jeanchastel-perf-budget (last updated 2026-10-24)