Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
git clone --depth 1 https://github.com/Kevin-Liu-01/LoopWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/rules/kevin-liu-01/loop/frontend-10x-system)<a href="https://agentmods.dev/rules/kevin-liu-01/loop/frontend-10x-system"><img src="https://agentmods.dev/badge/rules/kevin-liu-01/loop/frontend-10x-system/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/rules/kevin-liu-01/loop/frontend-10x-system"><img src="https://agentmods.dev/badge/rules/kevin-liu-01/loop/frontend-10x-system.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00790 | $0.00790 |
| Opus 5.5 | $0.00316 | $0.00316 |
| Sonnet 5.5 | $0.00158 | $0.00158 |
| Haiku 4.5 | $0.00079 | $0.00079 |
Grade A, and why
frontend-10x-system scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
What it actually says
High-Signal Frontend System
When generating or editing frontend code:
- Start by choosing an explicit art direction. Do not default to generic SaaS.
- Choose a methodology lens before choosing tactics: discovery, system, and evaluation.
- Pick one primary visual thesis: editorial technical, premium AI SaaS, cinematic 3D, architectural blueprint, neo-brutalist, or experimental.
- Build or extend a real design system before adding one-off styling.
- Default to Tailwind CSS v4
@themeplus runtime CSS variables for token transport. - Use tokens for color, spacing, radii, shadows, blur, z-index, and motion.
- Prefer OKLCH when authoring custom color scales.
- Reject raw hex colors, arbitrary spacing, and accidental radius/shadow drift.
- Prefer a type triad: display, sans, mono.
- Avoid Inter, Roboto, default shadcn aesthetics, and generic purple gradients unless the existing product already uses them or the user explicitly asks for them.
- Keep the primary UI to 3 active colors or fewer plus neutrals.
- Use authored line breaks in headlines.
- Use
text-wrap: balancefor headings and important short copy when helpful. - Use
antialiasedat the app layout level when it improves the chosen fonts. - Use
tabular-numsfor metrics, dates, prices, and changing numeric UI. - Make the product shot, dashboard mock, or diagram obey the same token system as the rest of the page.
- Keep navigation, CTA hierarchy, and proof blocks simple and obvious.
- Favor optical alignment over perfect geometry when they conflict.
- Default to bold but coherent decisions, not safe average ones.
- When the problem is craft quality, learn from manual-grade references like
userinterface.wiki, Rauno, Devouring Details, Interface Craft, and UI Playbook, not only from spectacle-heavy marketing sites. - Preserve concentric radii on nested surfaces: outer radius = inner radius + padding.
- Prefer subtle layered shadows over hard borders when the goal is depth.
- Give images and screenshots a low-opacity outline when they need a cleaner edge.
- CSS is the substrate for layout, tokens, and fallbacks. Do not default to raw CSS as the primary motion architecture.
- For React frontends, prefer Motion, GSAP, Lenis, R3F/drei, OGL, and React transitions where appropriate.
- If the product touches auth, user data, or public forms, include trust checks in the implementation: Supabase RLS where relevant, tested auth flows, rate limits, server-side validation, locked-down env vars, CAPTCHA on public forms when needed, CORS restrictions, and error handling that does not leak internals.
Composition rules:
- One hero focal point.
- One dominant CTA.
- One proof pattern near the claim.
- One repeated motif across the page.
Tactic rules:
- choose tactics intentionally: shell, density, motion, 3D/shaders, forms, data display, wait UX, resilience, search, AI, or docs
- do not cargo-cult patterns because they were trendy on one reference site
Ban these by default:
- centered-everything landing pages
- blur blobs with no material logic
- glassmorphism on white for no reason
- several competing accents
- five unrelated section styles on one page
- mismatched nested border radii
- important headlines left to accidental wrapping
- JS-only token systems when Tailwind v4 runtime vars would do the job better
If the task is docs-heavy, preserve docs usability first and add taste second.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 1mo ago First seen · 62 lines · 790 tokens per session scan A 8bef2a17a41a
frontend-10x-system is a cursor rule published in the GitHub repository Kevin-Liu-01/Loop (18 stars, last pushed 2d ago), licensed MIT. It adds 790 tokens to every session, about $0.0032 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other cursor rules, from other repositories
ui-craft-minimal
UI Craft minimal variant — Linear / Notion / Vercel whitespace-heavy UIs.
hatch3r-design-system-detection
Mandatory detection of existing design tokens, theme primitives, and component library before AI agents author new UI components.
hatch3r-ux-states-and-flows
Four-state surface contract (loading/empty/error/partial), user-flow decomposition before implementation, and microcopy + tone discipline for end-user UIs.
hatch3r-accessibility-standards
Accessibility standards covering WCAG 2.2 AA compliance, keyboard navigation, screen readers, and ARIA patterns.
hatch3r-component-conventions
Component structure, styling tokens, loading/error/empty states, form validation timing, and accessible label patterns for Vue, React, and JSX.
hatch3r-theming
Theming, dark mode, and color system conventions for the project.