ui-and-copy-discipline
Skill4 starsChanged 31 days agoMITView on GitHub
Frontend UI craftsmanship, typography, layout, copywriting, and WCAG 2.2 accessibility discipline. Enforces the 5 Craftsmanship Standards, Purpose Gates, and zero AI slop in web interfaces and marketing copy.
In this repository47 other files it gives its agents.
- AGENTS.mdAGENTS.md
- anti-slop-engineeringSkill
- anti-slop-engineeringSkill
- api-ergonomics-and-contract-governorSkill
- apple-craft-and-frame-budgetSkill
- apple-craft-and-interaction-physicsSkill
- asset-provenance-guardSkill
- backprop-guardSkill
- chaos-and-degradation-choreographerSkill
- chromatic-drift-guardSkill
- cognitive-load-and-attention-governorSkill
- comment-budget-guardSkill
- context-window-and-token-economics-architectSkill
- cryptographic-agility-and-secret-lifecycle-gateSkill
- decision-guardSkill
- demock-guardSkill
- design-system-drift-guardSkill
- deterministic-simulation-and-property-testingSkill
- distributed-consensus-and-concurrency-gateSkill
- finops-and-unit-economics-governorSkill
- future-scale-gateSkill
- grumpy-senior-staff-engineerSkill
- honeypot-trap-guardSkill
- house-tell-guardSkill
- install-anti-slopSkill
- llm-eval-and-prompt-injection-fortressSkill
- mandela-guardSkill
- mechanical-sympathy-and-perf-gateSkill
- mechanical-sympathy-profilerSkill
- motion-runtime-guardSkill
- observability-and-slo-architectSkill
- observability-and-slo-gateSkill
- paranoid-principal-architectSkill
- pr-boundary-guardSkill
- privacy-and-data-lineage-gateSkill
- re0-handoff-guardSkill
- reduction-guardSkill
- repo-memory-and-invariant-stewardSkill
- root-fix-gateSkill
- safe-evolution-and-rollout-choreographerSkill
- scientific-fault-isolationSkill
- styleseed-guardSkill
- supply-chain-and-sbom-sentinelSkill
- tagore-guardSkill
- type-level-and-ast-slop-gateSkill
- verification-oracle-architectSkill
- zero-downtime-migration-gateSkill
The file
We split this skill into the rules it gives an agent
6 rules. About 0 are common, 5 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈5
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 19
Before introducing any visual technique (gradient, backdrop-filter, border radius, box shadow, transition), answer:
- Distinctive: line 22
If the answer is "it looks like an AI default", "it makes the hero pop", or "it feels safe", the technique MUST be rejected or rewritten.
- Distinctive: line 23
If the answer identifies a distinct hierarchy elevation, interactive feedback, or content clarity, it stays.
- Very specific: line 30
Every visual and textual element must have a deliberate rationale.
- Distinctive: line 33
Every interactive element works or it does not exist.
- Distinctive: line 69
Content-Driven Breakpoints: Place responsive breakpoints where content naturally breaks or degrades readability, not based on arbitrary …
≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
UI & Copywriting Discipline
Enforce the 5 Craftsmanship Standards, Purpose Gates, and human accessibility across all user interfaces, components, and marketing prose.
1. The Core Purpose Test
Before introducing any visual technique (gradient, backdrop-filter, border radius, box shadow, transition), answer: What specific user or hierarchy goal does this technique serve?
If the answer is "it looks like an AI default", "it makes the hero pop", or "it feels safe", the technique MUST be rejected or rewritten. If the answer identifies a distinct hierarchy elevation, interactive feedback, or content clarity, it stays.
2. The 5 Craftsmanship Standards (C-1 to C-5)
C-1: Intentionality
Every visual and textual element must have a deliberate rationale. Never use default blue-to-purple gradients (from-blue-500 to-purple-500), ambient multi-layered blurred shadows, or generic hero radial glows without product design tokens.
C-2: Functional Completeness
Every interactive element works or it does not exist. A button, dropdown, search bar, or tab that renders but cannot execute its action is a defect, not decoration. Never ship non-functional mock UI elements.
C-3: Content-Driven Composition
Page layout and section order must flow directly from the product's actual domain data and user goals. Ban generic AI landing page templates: hero -> 3 uniform feature cards with icons -> fake testimonial carousel -> pricing grid -> FAQ accordion. Compose bespoke layouts that fit the data.
C-4: Resilience & Accessibility (WCAG 2.2 AA)
- Normal Text Contrast: $\ge 4.5:1$ against adjacent backgrounds.
- Large Text ($\ge 18\text{pt}$ or $\ge 14\text{pt}$ bold): $\ge 3.0:1$.
- UI Components & Graphical Edges: $\ge 3.0:1$ (WCAG 1.4.11).
- Touch Targets: Minimum $44 \times 44\text{px}$ on touch devices.
- Visible Focus States: Focus rings must have $\ge 3.0:1$ contrast against both the component and background. Never set
outline: nonewithout providing an accessible:focus-visiblereplacement.
C-5: Performance & Motion
- Animations must be hardware-accelerated (
transform,opacityonly). Never animate layout properties (width,height,margin,padding,top,left). - Always respect
prefers-reduced-motion: reduce. - Zero unconstrained requestAnimationFrame loops; terminate loops when animation values stabilize within $\epsilon \le 10^{-4}$.
3. The Three Rule Tiers
Tier 1: Hard Gates (Absolute Bans)
- Banned: Fake or unverified statistics in marketing copy ("Trusted by 10,000+ teams" when fictitious).
- Banned: AI buzzword inflation ("Unleash the power of AI", "Revolutionize your workflow", "Next-generation AI-powered platform").
- Banned: Unrequested placeholder image/asset creation.
- Banned: Decorative separator comments in source code (
// =======================,/* ---- ROUTES ---- */). - Banned: Low-contrast text failing WCAG 2.2 AA standards.
- Banned: Dev-notes or mock disclaimers in user-visible DOM/JSX (
"Note: in production this connects to backend").
Tier 2: Purpose Gates (Permitted ONLY with Documented Justification)
- Gradients: Permitted only when establishing intentional hierarchy between primary and secondary surfaces.
- Glassmorphism (
backdrop-filter: blur(...)): Dose cap of at most 1–2 elements per viewport (e.g. sticky navigation bar only). - Pill Border Radii (
rounded-full): Permitted on compact status chips and floating action buttons; banned on text inputs, textareas, and content cards. - Micro-Animations: Permitted when communicating state changes, transitions, or feedback; banned as gratuitous perpetual distraction.
Tier 3: Quality Locks (Responsive & Mobile Discipline)
- Content-Driven Breakpoints: Place responsive breakpoints where content naturally breaks or degrades readability, not based on arbitrary device spec sheets.
- Mobile-First Layout: Mobile viewports must be deliberately designed, not desktop layouts forcibly squeezed into narrow widths.
- Safe Area Insets: Support iOS/Android display cutouts (
env(safe-area-inset-top),env(safe-area-inset-bottom)).
4. Automated Verification Commands
# Verify color contrast pairing:
bunx anti-slop-engineering contrast "#FFFFFF" "#1E293B"
# Scan workspace stylesheets and components for contrast violations:
bunx anti-slop-engineering contrast --scan src/
# Run single Diff Guard triage (narrow diff check):
bunx anti-slop-engineering diff --base HEAD
# Run applicable multi-guard audit scan (Diff, SRE, Contrast, etc.):
bunx anti-slop-engineering audit --base HEADDoes 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
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.

