By realdeal88
A complete framework for making and judging great design — eight lenses (Purpose, Agency, Responsibility, Familiarity, Flexibility, Simplicity, Craft, Delight) that turn a vague 'make it good' into concrete, defensible decisions. Auto-invokes whenever design quality matters on any platform — web, mobile, desktop, product: building, redesigning, or critiquing any UI, flow, or component, plus design decisions from feature-scoping to permission prompts to accessibility.
Based on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Eight lenses for great design — an auto-invoked decision-and-critique framework for Claude Code.
Turns a vague "make it good" into concrete, defensible decisions. Fires whenever design quality matters — on any platform, even when nobody says the word "design."
"Make it look good" is the least actionable instruction in software. It has no edges — nothing to build toward, nothing to check against. So an agent guesses, you correct, it guesses again.
design-lens replaces the guess with a frame. Eight lenses turn the fuzzy ask into a set of working questions with right answers: What is this for, and what can be cut? Can people undo it? Am I asking for that permission at the right moment? Does this detail say someone cared? Each lens is a place to look, a failure mode to avoid, and a line you can defend in review.
It's a skill, so it loads itself. The moment you're shaping how something looks, behaves, or feels to a person, Claude pulls the lenses in — no command to remember, no principle to recite.
| # | Lens | The question it forces |
|---|---|---|
| 1 | Purpose | What is this for, and what can I cut? Every element spends attention — make each earn it. |
| 2 | Agency | Can people work their way? Is every action undoable, every destructive one confirmed? |
| 3 | Responsibility | Am I acting in this person's interest — asking for data in context, guarding against misuse? |
| 4 | Familiarity | Am I building on what people already know, instead of reinventing the obvious? |
| 5 | Flexibility | Does it fit real, varied use — across contexts, devices, and abilities? |
| 6 | Simplicity | Is there exactly enough? (Simple ≠ minimal — hiding everything isn't simple.) |
| 7 | Craft | Does every detail say someone cared? Polish builds trust; jank quietly destroys it. |
| 8 | Delight | What emotion is this creating — and does the whole design reinforce it? |
They aren't a scorecard. There's no formula — leaning into one often costs another (simple vs. flexible, freedom vs. safety). Naming that trade-off and choosing on purpose is the design work.
# In Claude Code:
/plugin marketplace add realdeal88/design-lens
/plugin install design-lens
That's it. The skill auto-invokes from then on. To confirm it's loaded, ask Claude to "review this screen against the design lenses" — it should run the eight-lens self-audit.
Copy the skill into your skills directory:
git clone https://github.com/realdeal88/design-lens
cp -r design-lens/skills/design-lens ~/.claude/skills/
design-lens/
├── skills/design-lens/
│ ├── SKILL.md # the working tool — eight lenses, usage modes, self-audit
│ └── references/
│ └── design-principles.md # the deep reference — each lens expanded with failure modes
└── .claude-plugin/
├── plugin.json # plugin manifest
└── marketplace.json # single-repo marketplace
SKILL.md is what loads on trigger. references/design-principles.md is pulled only when a
decision needs more than the one-line lens — keeping the always-on footprint small.
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimnpx claudepluginhub realdeal88/design-lens --plugin design-lensیک ایدهٔ خام را به یک پرامپت اجرایی و خودراستیآزما تبدیل کن — برگرفته از مستندات مهندسی پرامپت خود Anthropic. هر پرامپت فورجشده یک معیار موفقیت قابلراستیآزمایی دارد تا اجرا خودش تا «انجامشده» پیش برود.
Your project's always-on memory. A set of living files (goals · tracker · decisions · mistakes · log) that auto-catch-up at session start and auto-snapshot at every compaction and session end — so no context is ever lost to a /compact or a restart. Runs itself through five non-blocking hooks; never coerces work.
Claude Code için her zaman açık proje hafızası. Oturum başında auto-catchup, her compaction ve oturum sonunda auto-snapshot yapan canlı dosyalar — böylece hiçbir bağlam sıfırlamayla kaybolmaz. Beş bloke etmeyen hook ile kendi kendine çalışır; işi asla zorlamaz.
Forge a rough idea into an execution-ready prompt. Distilled from Anthropic's own prompt-engineering documentation — every forged prompt carries a verifiable success criterion so the run drives itself to done.
حافظهٔ همیشهفعال پروژه برای Claude Code. فایلهای زندهای که هنگام شروع نشست auto-catchup و در هر compaction و پایان نشست auto-snapshot میکنند — تا هیچ کانتکستی با ریست از دست نرود. با پنج هوک غیرمسدودکننده خودش کار میکند؛ هرگز کار را اجبار نمیکند.
Comprehensive skill pack with 66 specialized skills for full-stack developers: 12 language experts (Python, TypeScript, Go, Rust, C++, Swift, Kotlin, C#, PHP, Java, SQL, JavaScript), 10 backend frameworks, 6 frontend/mobile, plus infrastructure, DevOps, security, and testing. Features progressive disclosure architecture for 50% faster loading.
UI/UX design intelligence. 84 styles, 161 palettes, 73 font pairings, 25 charts, 17 stacks (React, Next.js, Vue, Nuxt.js, Nuxt UI, Svelte, Astro, SwiftUI, React Native, Flutter, Tailwind, shadcn/ui, Jetpack Compose, Angular, Laravel, JavaFX, Three.js). Actions: plan, build, create, design, implement, review, fix, improve, optimize, enhance, refactor, check UI/UX code. Projects: website, landing page, dashboard, admin panel, e-commerce, SaaS, portfolio, blog, mobile app. Elements: button, modal, navbar, sidebar, card, table, form, chart. Styles: glassmorphism, claymorphism, minimalism, brutalism, neumorphism, bento grid, dark mode, responsive, skeuomorphism, flat design. Topics: color palette, accessibility, animation, layout, typography, font pairing, spacing, hover, shadow, gradient.
Develop, test, build, and deploy Godot 4.x games with Claude Code. Includes GdUnit4 testing, web/desktop exports, CI/CD pipelines, and deployment to Vercel/GitHub Pages/itch.io.
A growing collection of Claude-compatible academic workflow bundles. Covers scientific figures, manuscript writing and polishing, reviewer assessment, citation retrieval, data availability, paper reading, literature search, response letters, paper-to-PPTX conversion, and evidence-grounded Chinese invention patent drafting. Rules are organized as reusable skill folders with explicit workflows and quality checks.
Comprehensive PR review agents specializing in comments, tests, error handling, type design, code quality, and code simplification
Comprehensive feature development workflow with specialized agents for codebase exploration, architecture design, and quality review