stark
Use for UI/UX design, product flows, frontend or native interface design, design audits, design translation, asset planning, motion direction, design tokens, or distinctive non-generic apps and websites.
Install / Use
npx skills add hashgraph-online/awesome-codex-plugins --skill starkInstalls into whichever agent you are using.
SKILL.md
Installable skill definition
Quality Score
Category
DesignSupported Platforms
Our assessment of stark
stark scores 84/100 on our quality scale, 193rd of 260 Design skills we index.
Its SKILL.md is 8.7 KB long, well organised into 17 sections and no code examples: a thorough specification that gives an agent plenty to work with.
With 1,075 GitHub stars, it is one of the more widely adopted skills in the catalogue.
Maintenance, license and trust
- The repository was last updated about 2 months ago, so stark is actively maintained.
- Our last check on 2026-11-05 found the source still online.
- It is released under the Apache-2.0 license, a permissive license that allows use, modification and commercial use with attribution.
- Its trust signals score 100/100, with no cautions. These come from repository metadata, not a code audit — read the skill file before letting an agent act on it.
Safety scan
No issues foundOur scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands. An AI review of the same text found nothing harmful.
AI review by kimi-k2.7-code on 2026-10-07. Automated pattern scan on 2026-10-07. It catches known dangerous patterns, not every risk — read a skill before letting an agent act on it.
stark compared with similar skills
All 4 of these similar skills score higher than stark; compare them before choosing.
| Skill | Score | Stars | Updated | Format |
|---|---|---|---|---|
| stark (this skill)by hashgraph-online | 84 | 1.1k | 53d ago | SKILL.md |
| taste-skillby Leonxlnx | 100 | 116.8k | 8d ago | CLAUDE.md |
| Claude-Code-Game-Studiosby Donchitos | 100 | 39.7k | 7d ago | CLAUDE.md |
| x402by browser-use | 100 | 121.0k | 3d ago | SKILL.md |
| algorithmic-artby anthropics | 100 | 177.9k | 55d ago | SKILL.md |
Frequently asked questions
- How do I install stark?
- Run
npx skills add hashgraph-online/awesome-codex-plugins --skill stark. The install tabs above show the steps for each supported agent. - Which AI agents does stark work with?
- It is written for Universal, as a SKILL.md file. Other agents that read the same format can often use it too.
- Is stark safe to use?
- Our scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands. An AI review of the same text found nothing harmful. It is Apache-2.0-licensed and scores 100/100 on trust signals. Skills are instructions an agent will follow, so read the file before installing it and do not approve commands you do not understand.
- Is stark still maintained?
- The repository was last updated about 2 months ago, so stark is actively maintained.
Skill content
View source on GitHubname: stark description: Use for UI/UX design, product flows, frontend or native interface design, design audits, design translation, asset planning, motion direction, design tokens, or distinctive non-generic apps and websites. Covers web and React, Windows and WinUI/Mica, Apple and SwiftUI, Android and Jetpack Compose/Material, cross-platform UI, and token export. Route to the focused skill, then load only the references required for the selected surface and failure mode.
Stark
Stark is a design router for AI coding agents. Use it to make product, platform, visual, interaction, and quality decisions before implementation.
Operating rule
Keep context proportional to the task:
- Select no more than two routed skills.
- Load two core references for the selected surface.
- Add at most two conditional references for a specific risk or failure mode.
- Load post-render gates only after rendered evidence exists.
Do not read the entire reference library. Use references/ui-patterns/README.md, references/ux-patterns/README.md, and references/web-patterns/README.md as catalogs when the tables below do not cover the request.
Route the request
| Request | Routed skill |
|---|---|
| Web UI, landing page, dashboard, React, Next, Astro, Tailwind | skills/web-design/SKILL.md |
| Product flow, onboarding, checkout, IA, usability, forms | skills/ux-design/SKILL.md |
| Windows, WinUI, WPF, Fluent | skills/windows-design/SKILL.md |
| Apple, iOS, iPadOS, macOS, SwiftUI | skills/apple-design/SKILL.md |
| Android, Material, Jetpack Compose | skills/android-design/SKILL.md |
| Tauri, Electron, React Native, Flutter, CMP, shared UI | skills/cross-platform-design/SKILL.md |
| Design tokens, themes, token export | skills/design-tokens/SKILL.md |
| Platform is genuinely unclear | skills/design-router/SKILL.md |
Combine UX with one platform skill when the request includes both workflow and implementation. Preserve the UX brief when platform guidance is applied.
Use scripts/detect_platform.py only as a deterministic fallback. Treat ambiguous as a request for one short clarification, not permission to default to web.
Select a reference bundle
Choose one primary bundle. Read its two core references. Add conditional references only when their trigger is present.
Product application
Core:
references/ux-patterns/task-ergonomics.mdreferences/ui-patterns/interaction-state-matrix.md
Conditional:
- Responsive or multi-viewport:
references/ui-patterns/responsive-adaptation-gate.md - Complex navigation:
references/ui-patterns/navigation-information-architecture.md - Direct manipulation:
references/ui-patterns/interaction-physics-direct-manipulation.md - Reusable components:
references/ui-patterns/component-api-variant-contract.md
Landing page or product proof
Core:
references/ui-patterns/creative-direction.mdreferences/ui-patterns/page-proof-architecture.md
Conditional:
- Conversion or pricing:
references/ui-patterns/conversion-proof-system.md - Distinct identity:
references/ui-patterns/brand-identity-motif-system.md - Product media:
references/ui-patterns/asset-realism-matrix.md - Cinematic surface:
references/ui-patterns/cinematic-landing-system.md
Animation-led surface
Core:
references/ui-patterns/animation-creation.mdreferences/ui-patterns/choreography-state-machine.md
Conditional:
- Library selection:
references/ui-patterns/motion-library-playbooks.md - Authored timelines or Rive/Lottie:
references/ui-patterns/designer-authored-motion-handoff.md - Heavy runtime:
references/ui-patterns/performance-budget-contract.md - Acceptance frames:
references/ui-patterns/motion-frame-qa-contract.md
Dashboard or analytics
Core:
references/ui-patterns/dashboard-insight-hierarchy.mdreferences/ui-patterns/data-visualization-library-selection.md
Conditional:
- Realistic fixtures:
references/ui-patterns/product-data-content-system.md - Dense repeated work:
references/ux-patterns/task-ergonomics.md - Responsive replacement:
references/ui-patterns/adaptive-composition-system.md - Data states:
references/ui-patterns/interaction-state-matrix.md
Design system or reusable library
Core:
references/ui-patterns/design-system-production-loop.mdreferences/ui-patterns/token-implementation-contract.md
Conditional:
- Public component API:
references/ui-patterns/component-api-variant-contract.md - State coverage:
references/ui-patterns/component-state-gallery.md - Theme and density:
references/ui-patterns/theme-mode-density-system.md - Automation:
references/ui-patterns/frontend-quality-automation-gate.md
Audit or repair
Core:
references/ui-patterns/ui-audit-rubric.mdreferences/ui-patterns/visual-repair-playbook.md
Conditional:
- Usability concern:
references/ux-patterns/usability-heuristic-evaluation.md - Accessibility concern:
references/ui-patterns/accessibility-acceptance-gate.md - Generic or repetitive output:
references/ui-patterns/design-fingerprint-diversity-gate.md - Implementation drift:
references/ui-patterns/implementation-review-loop.md
Research or reference-led work
Core:
references/ui-patterns/reference-board-workflow.mdreferences/ui-patterns/research-synthesis-contract.md
Conditional:
- Compare against a stronger example:
references/ui-patterns/benchmark-repair-loop.md - Avoid copying identity:
references/ui-patterns/reference-analysis.md
First-run or risky flow
Core:
references/ux-patterns/first-run-empty-state-system.mdreferences/ux-patterns/task-flow-acceptance-harness.md
Conditional:
- Forms and validation:
references/ui-patterns/form-state-validation-system.md - Checkout or trust:
references/ui-patterns/conversion-proof-system.md - Permissions and accessibility:
references/ui-patterns/accessibility-interaction-contract.md
Produce a compact decision brief
Before code, state:
- product job and primary user;
- target platform and implementation track;
- primary object and action;
- selected structural pattern;
- visual direction and one product-specific motif;
- required states and recovery path;
- responsive or native adaptation;
- motion level and library owner, if any;
- selected references and why they apply;
- acceptance evidence.
Keep this brief actionable. Do not turn it into a design essay.
Implement and validate in stages
- Define the product object, primary action, states, and recovery.
- Choose platform idiom and structural pattern.
- Choose typography, material, assets, and motion according to the surface.
- Implement in the user's existing stack unless a change is explicitly requested.
- Inspect the rendered result.
- Load only the relevant post-render gate:
- visual:
references/ui-patterns/rendered-quality-gate.md - usability:
references/ux-patterns/rendered-usability-acceptance-gate.md - typography:
references/ui-patterns/rendered-typography-quality-gate.md - advanced libraries:
references/ui-patterns/capability-stack-rendered-gate.md
- visual:
- Apply the highest-impact repair and re-check the failed evidence.
Commands
Use command briefs when the user explicitly asks for one of these modes:
| Mode | Command |
|---|---|
| General design | commands/stark.md |
| Audit | commands/stark-audit.md |
| Full design direction | commands/stark-director.md |
| Quality scoring | commands/stark-quality.md |
| Browser/runtime QA | commands/stark-qa.md |
| Release-quality gate | commands/stark-quality-gate.md |
| Usability | commands/stark-usability.md |
| Animation | commands/stark-animation.md |
| Assets | commands/stark-assets.md |
| Reference research | commands/stark-reference-board.md |
| Second attempt | commands/stark-regenerate.md |
| Design translation | commands/stark-translate.md |
| Dogfood evaluation | commands/stark-dogfood.md |
Helper scripts
- Use
scripts/detect_platform.pyonly when semantic platform routing remains ambiguous. - Use
scripts/token_export.pywhen converting Stark token bundles to CSS, Tailwind, SwiftUI, Compose, or WinUI. - Maintainers use
scripts/validate_repository.pyandscripts/build_package.py; they are not part of normal design execution.
Guardrails
- Prefer product-specific structure over generic cards, gradients, and slogans.
- Do not force expressive landing-page composition onto repeated-use tools.
- Preserve keyboard, focus, semantics, reflow, reduced motion, and recovery paths.
- Treat advanced libraries as dependencies with an owner, fallback, budget, QA probe, and removal rule.
- Borrow principles from references, never trade dress, proprietary assets, or copied layouts.
- Do not claim rendered quality without rendered evidence.
- Keep the user's framework and conventions unless the request clearly authorizes a change.
Related Skills
taste-skill
116.8kAgent skills that give coding agents taste: art direction, reference images and a browser check instead of generic AI pages
Claude-Code-Game-Studios
39.7kTurn Claude Code into a full game dev studio — 49 AI agents, 72 workflow skills, and a complete coordination system mirroring real studio hierarchy.
x402
121.0kSet up Browser Use Cloud payments with x402 — pay per request from a crypto wallet (USDC on Base mainnet), no signup or API key.
algorithmic-art
177.9kCreating algorithmic art using p5.js with seeded randomness and interactive parameter exploration. Use this when users request creating art using code, generative art, algorithmic art, flow fields, or particle systems.
Languages
Trust signals
From repository metadata: license, adoption, age and documentation. Not a code audit — see the Safety scan above for what the skill file itself contains.
