agentleFS
boazsegevboazsegev/omoya/skills/web-frontend/SKILL.md

web-frontend

Skill1 starsChanged 6 days agoMITView on GitHub

This skill should be used when writing or reviewing HTML, CSS, browser JavaScript, SEO or installable web apps.

We split this skill into the rules it gives an agent

7 rules. 0 are common, 4 are distinctive, and 3 are very specific to this project.

  • 0

    Common

    Most of the 8 edited versions we read keep these.

  • 4

    Distinctive

    This author's own choices. Worth a look.

  • 3

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Prefer platform features to code, and code to dependencies: HTML/CSS > JavaScript > Frameworks

    line 10
  2. Distinctive:

    Use semantic elements first.

    line 16
  3. Distinctive:

    Use logical properties and intrinsic layout before breakpoints.

    line 22
  4. Very specific:

    Keep specificity low with @layer.

    line 23
  5. Distinctive:

    Treat external data as untrusted: insert text, sanitize HTML, allowlist URL schemes.

    line 31
  6. Very specific:

    Use a strict CSP without unsafe-inline.

    line 32
  7. Very specific:

    Avoid storing non-trivial data as plaintext.

    line 33

“Common” is measured: kept by at least 6 in 10 of the 8 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Frontend Practices

Architecture

  • Separate content from style and behavior.
  • Prefer platform features to code, and code to dependencies: HTML/CSS > JavaScript > Frameworks
  • Drive behavior from markup and data (data-*, delegated events, lookup tables), not per-element code.
  • Render from one state; never read state back from the DOM.
  • Enhance progressively: content and navigation work without JS.

HTML and Accessibility

  • Use semantic elements first; add ARIA only for gaps.
  • Meet WCAG 2.2 AA with full keyboard and screen-reader operation.
  • Give every control an accessible name; set lang/dir; manage focus on dynamic change.

CSS

  • Tokenize design values; components use semantic tokens.
  • Use logical properties and intrinsic layout before breakpoints.
  • Keep specificity low with @layer; honor color scheme, reduced motion and print.

Performance

  • Meet Core Web Vitals: LCP <2.5 s, INP <200 ms, CLS <0.1.
  • Reserve media space; prioritize the LCP resource; defer the rest.
  • Batch DOM reads before writes; release listeners and observers.

Security

  • Treat external data as untrusted: insert text, sanitize HTML, allowlist URL schemes.
  • Use a strict CSP without unsafe-inline; keep secrets server-side.
  • Avoid storing non-trivial data as plaintext. Use encryption when available. Use the safe-storage.js resource when storing non-trivial information in localStorage.

Installable Apps

  • Serve over HTTPS with a manifest (name, start_url, scope, display: standalone, colors, 192/512 and maskable icons) plus apple-touch-icon.
  • Work offline via a versioned network-first service worker: copy the sw.js resource and edit its constants; it purges only app's old caches.
  • Design for standalone launch: in-app navigation, safe-area insets (viewport-fit=cover).

Discoverability

  • One intent per page: unique title, one h1, answer-first sections, canonical URL.
  • Ship content in initial HTML; JSON-LD only for visible facts; hreflang per language URL.
  • Serve AI agents: choose search and training crawler access separately in robots.txt; offer /llms.txt and Markdown views; keep actions labeled and operable without vision.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.