14kb-club

Audit web projects against the 14KB TCP slow start budget. Conservative approach: report and suggest, never auto-apply.

Safety Notice

This listing is imported from skills.sh public index metadata. Review upstream SKILL.md and repository scripts before running.

Copy this and send it to your AI assistant to learn

Install skill "14kb-club" with this command: npx skills add handxr/14kb-club/handxr-14kb-club-14kb-club

14KB Club

Audit web projects against the 14KB TCP slow start budget. Conservative approach: report and suggest, never auto-apply.

Workflow

  • Identify target directory or file

  • Run the auditor script: python3 <skill-path>/scripts/14kb_club.py <file_or_dir> --json

  • Parse JSON output

  • Present budget bar (if build detected) + issues grouped by severity

  • Propose fixes starting with errors, then warnings

  • Wait for user approval before applying changes

  • Re-run after fixes to verify improvement

Running the Auditor

Single file

python3 <skill-path>/scripts/14kb_club.py index.html

Entire project (auto-detects framework + build folder)

python3 <skill-path>/scripts/14kb_club.py src/

Scan build output for real gzipped sizes

python3 <skill-path>/scripts/14kb_club.py dist/

Custom budget (default: 14KB)

python3 <skill-path>/scripts/14kb_club.py . --budget 10

Force a specific build directory

python3 <skill-path>/scripts/14kb_club.py . --build-dir .next/

Include low-severity hints

python3 <skill-path>/scripts/14kb_club.py . --strict

Machine-readable JSON

python3 <skill-path>/scripts/14kb_club.py . --json

What It Detects

CSS Critical Path (severity: error)

  • External CSS in <head> without media attribute — render-blocking

  • No inline <style> for critical CSS — all CSS is external

  • @import chains in CSS files — sequential blocking requests

  • CSS files > 14KB gzipped

JS Blocking (severity: error)

  • <script src> without defer /async in <head> — blocks parsing

  • Known third-party blocking scripts (GA, GTM, Facebook, Intercom, etc.)

  • Legacy DOM write methods — blocks parsing

  • JS files > 14KB gzipped

Fonts (severity: warning)

  • @font-face without font-display: swap — invisible text (FOIT)

  • Font files not preloaded — late discovery

  • Google Fonts loading 4+ weights — each adds 20-50KB

  • More than 2 font families

HTML Bloat (severity: warning)

  • Inline SVGs > 2KB

  • data: URIs > 1KB

  • HTML > 14KB uncompressed

  • Excess meta/preload tags (10+)

  • Inline style blocks > 5KB

Build Mode

When a build folder is detected (or specified with --build-dir ), the script measures real gzipped sizes and renders a budget bar:

Budget: [===========> ] 11.2 / 14.0 KB gzipped (80%)

HTML: 3.2 KB [==== ] 23% CSS: 4.1 KB [===== ] 29% JS: 3.9 KB [===== ] 28% Fonts: 0.0 KB [ ] 0%

Status: PASS — 2.8KB under budget

Auto-detected frameworks: Next.js, Vite, Astro, Nuxt, Angular, SvelteKit, Remix, Gatsby, CRA, Plain HTML, PHP.

Interpreting Results

  • !! Error: Render-blocking issue — fix immediately

  • !~ Warning: Significant weight contributor — fix in performance-critical pages

  • ~~ Info: Minor optimization opportunity (only with --strict )

Fix Priority

  • Render-blocking CSS — inline critical CSS, defer the rest

  • Render-blocking JS — add defer or async , move before </body>

  • Third-party scripts — defer/async or load via tag manager

  • Font loading — add font-display: swap

  • preload
  • HTML bloat — extract SVGs, remove data URIs, trim meta tags

Detailed Rules

For complete detection rules, edge cases, and fix patterns, read references/14kb-rules.md.

Source Transparency

This detail page is rendered from real SKILL.md content. Trust labels are metadata-based hints, not a safety guarantee.

Related Skills

Related by shared tags or category signals.

Security

tailwind-conflict-auditor

No summary provided by upstream source.

Repository SourceNeeds Review
General

philosophy-software-design

No summary provided by upstream source.

Repository SourceNeeds Review
Security

compliance-evidence-assembler

把审计所需证据整理成目录、清单和缺失项,便于后续评审。;use for compliance, evidence, audit workflows;do not use for 伪造证据, 替代正式审计结论.

Archived SourceRecently Updated
Security

skillguard-hardened

Security guard for OpenClaw skills, developed and maintained by rose北港(小红帽 / 猫猫帽帽). Audits installed or incoming skills with local rules plus Zenmux AI intent review, then recommends pass, warn, block, or quarantine.

Archived SourceRecently Updated