mobile-ui-ux
Measurable usability and interaction rules for iPhone apps, mobile web, and mobile games, with a 7-category scoring rubric, an independent-judge prompt, and a…
Skill by cskwork
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/cskwork-mobile-ui-ux
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: spreadsheets, design, testing, agent-building
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/mobile-ui-ux
What's inside
- Skill name
mobile-ui-ux- Description
- Measurable usability and interaction rules for iPhone apps, mobile web, and mobile games, with a 7-category scoring rubric, an independent-judge prompt, and a pre-ship checklist. Use for any UI/UX design, review, or implementation work on an iPhone app, a mobile web page or PWA, or a mobile game (HUD, combat readability, drag-and-drop, onboarding, settings sheets). Also use when asked to score, judge, or QA a mobile interface, or to set up project UI/UX overrides. Pairs with impeccable, which owns aesthetic direction.
- License
- MIT
- Instructions
- 131 lines in SKILL.md · view on GitHub
Files (18)
- CUSTOMIZE.md
- SKILL.md
- checklist.md
- references.md
- rubric.md
- accessibility.md · rules/
- color.md · rules/
- feedback-motion.md · rules/
- forms.md · rules/
- games.md · rules/
- layout.md · rules/
- localization.md · rules/
- navigation.md · rules/
- performance.md · rules/
- states.md · rules/
- touch.md · rules/
- typography.md · rules/
- web.md · rules/
SKILL.md (first lines)
# mobile-ui-ux Concrete rules for usable phone interfaces: sizes in pt, contrast ratios, durations, layout limits. Each rule has an ID (`T-3`, `G-12`) so reviews, overrides, and commit messages can point at it. This skill answers "can a person on a small phone read it, reach it, understand it, and recover from mistakes?" It does not pick a visual style. For aesthetic direction, typography choice, palette, and critique of taste, use **impeccable**; then run this skill's rules and rubric on the result. When the app is SwiftUI, pair with **swiftui-pro** for code-level API use (deprecated APIs, data flow, navigation, the SwiftUI way to meet a rule). Where they disagree, this skill's floors win; for example, LOC-1 allows a UIKit-backed label even though swiftui-pro avoids UIKit by default. ## When to use - Designing or building any screen for an iPhone app, a mobile web page, or a mobile game. - Reviewing screenshots or a running build for UI/UX quality. - Scoring a build against a quality bar, or writing a prompt for an independent judge. - Creating or editing a project's UI/UX overrides file. Not for: backend work, store metadata copy, App Review compliance (see the workspace's release docs). ## Files | File | Read when | |---|---| | `rules/layout.md` | Any screen. Safe areas, margins, grids, small-device fit, overflow. | | `rules/touch.md` | Anything tappable or draggable. Target sizes, spacing, thumb reach, gestures. | | `rules/typography.md` | Any text. Sizes, Dynamic Type, line length, truncation. | | `rules/color.md` | Any color or contrast choice, dark mode, color-blind safety. | | `rules/feedback-motion.md` | Press states, animation timing, haptics, sound, Reduce Motion. | | `rules/navigation.md` | Screen structure, tab bars, sheets, back, destructive actions. | | `rules/forms.md` | Text input, pickers, validation, keyboards. | | `rules/states.md` | Empty, loading, error, offline, first-run states. | | `rules/accessibility.md` | VoiceOver, Dynamic Type at large sizes, alternatives to gestures. | | `rules/localization.md` | More than one language, or any Korean/CJK text. | | `rules/performance.md` | Load time, input latency, frame rate as felt by the user. | | `rules/web.md` | Mobile web or PWA: viewport, `env()`, `dvh`, input zoom. | | `rules/games.md` | Any game: HUD, combat readability, damage numbers, drag-and-drop, FTUE. | | `rubric.md` | Scoring a build; judge prompt template. | | `checklist.md` | Before calling a UI done, before a release commit. | | `CUSTOMIZE.md` | Setting up or reading project overrides. | | `references.md` | Source for any rule; cite it when a rule is challenged. | Load only the rule files the task touches. A settings screen in an app needs layout, touch, typography, color, navigation, forms. A game battle screen needs layout, touch, typography, color, feedback-motion, games. ## Workflow ### 1. Read project overrides Look in the project root, in this order, and use the first that exists: 1. `.ui-ux.md` 2. `docs/ui-ux-overrides.md` 3. `DESIGN.md` section headed `## UI/UX overrides` The file sets platform profile, smallest and largest supported device, languages, bar threshold, genre references, brand constraints, and rule exceptions. Format and defaults: `CUSTOMIZE.md`. If no file exists, use the defaults below and say so in one line of your output. Defaults when no override exists: - Smallest device: iPhone SE (3rd gen), 375 x 667 pt, @2x (1334 x 750 px landscape). - Largest device: iPhone Pro Max class, 440 x 956 pt, @3x. - Languages: the languages the project already ships. If it ships Korean, Korean is primary. - Bar: every rubric category >= 4, average >= 4.3. - Appearance: light and dark if the app supports both; games: whatever the game ships. An override can relax a rule only with a stated reason. It can never relax rules marked **(floor)**: those are accessibility and legibility minimums. ### 2. Pick the platform profile | Profile | Use for | Rule files always loaded | |---|---|---| | `app` | Native iPhone app (SwiftUI, UIKit, React Native, Flutter) | layout, touch, typography, color, navigation, states, accessibility | | `web` | Mobile web page, web app, PWA | the `app` set + `web`, `performance` | | `game` | Any game, any engine (Godot, Unity, SpriteKit, web canvas) | layout, touch, typography, color, feedback-motion, games, accessibility | A game's menus, shop, and settings are app-like screens: apply navigation and forms rules there too. ### 3. Apply the rules while designing - Design for the smallest device first. Fit there, then let the layout breathe on large devices. Never design at Pro Max size and shrink. - Before placing any element, decide its priority (primary, secondary, tertiary). Primary content gets the center and the largest size; tertiary content can move behind a tap. - Every tap target >= 44 x 44 pt (T-1). Every text >= 11 pt at the smallest device (TY-1). Both measured in points on the real device, not in canvas pixels. - When something does not fit, follow L-8: show less and add an overflow cue, never shrink below floors or clip silently. - Write the rule ID next to any decision that departs from a default, in code comments or the design note. ### 4. Self-check with the checklist Walk `checklist.md` for the profile. Mark each item pass, fail, or n/a with the evidence (screenshot name, test name). A fail is fixed before scoring, not explained away. ### 5. Score with the rubric and iterate - Score each of the 7 categories in `rubric.md` from 1 to 5 using its anchors. - If any category is under the bar, list fixes in order of impact (largest score gain per effort first), fix them, recapture, rescore. - For release candidates or when the owner asks for an independent review, send the judge prompt in `rubric.md` to a separate agent that has not seen the code or your reasoning: only screenshots, the rubric, and the bar. Your own score does not count as the gate. - Stop when the bar is met. Do not keep polishing past the bar unless asked. ### 6. Capture evidence Every review round produces screenshots, named `<screen>_<device>_<lang>_<state>.png`, for example `battle_se3_ko_mid_combat.png`. Minimum set per round: - Each primary screen at the smallest and the largest supported device. - Each supported language on the screen with the most text. - The states that rules call out: empty, loading, error, full/limit-reached, mid-drag, settings open. - Games: prep, mid-combat, result, and one frame strip (3 to 5 frames) of the busiest transition. - Accessibility pass: one shot at the largest Dynamic Type size the app supports (apps) and one with Reduce Motion on, if motion changes. Record what was captured and on which build in the project's QA doc so the next round does not repeat it. ## Composing with impeccable | Question | Skill | |---|---| | What should this feel like? Which typeface, palette, layout concept? | impeccable | | Is this bold enough, too generic, on-brand? | impeccable | | Can a person reach, read, and operate it on an SE? Is the HP bar readable? | mobile-ui-ux | | Did it pass the bar? | mobile-ui-ux rubric | Typical order: impeccable sets direction and builds; mobile-ui-ux rules constrain the build; impeccable polishes; mobile-ui-ux scores. If the two disagree, rules marked **(floor)** win; otherwise the project override decides; otherwise impeccable owns look and this skill owns usability. ## Output format for reviews
From cskwork/skills, MIT.
More from cskwork/skills
- 0.586score
motion-graphics
cskwork
Make code-driven motion-graphics videos rendered from HTML, frame-accurate and reproducible.
SkillAFreshness: A★ 3Claude Code +9 - 0.586score
db-intelligence
cskwork
Unified database intelligence for PostgreSQL, MySQL, SQLite, and MongoDB in one skill: detect the engine, connect credential-safe, extract schema, build an…
SkillAFreshness: A★ 3Claude Code +9 - 0.586score
diagram-design
cskwork
Create branded architecture, architecture delta, IT current-state, flowchart, sequence, state machine, ER/data model, timeline, swimlane, quadrant…
SkillAFreshness: A★ 3Claude Code +9 - 0.586score
gpt-image-2
cskwork
Generate images with ChatGPT's current image model (GPT Image 2 at the time of writing, and whatever model succeeds it) inside your agent, using your existing…
SkillAFreshness: A★ 3Claude Code +9 - 0.586score
image-first-ui
cskwork
Image-first interface creation.
SkillAFreshness: A★ 3Claude Code +9 - 0.586score
bughunt
cskwork
Find and fix interaction bugs in games, CRUD apps, AI chat and media apps with seeded self-playing personas, per-step invariant oracles, reproducible failure…
SkillAFreshness: A★ 3Claude Code +9 - 0.581score
quality-loop
cskwork
Product quality bar and improvement loop for iPhone apps, websites, web apps, games, and other software.
SkillAFreshness: A★ 3Claude Code +9 - 0.581score
sdlc-kit
cskwork
Four-phase SDLC loop for any codebase: intent (90% clear: explore, reproduce, research, ask the requester), plan (parallel scouts, one PLAN.md), build…
SkillAFreshness: A★ 3Claude Code +9 - 0.580score
jk
cskwork
Manage Jenkins controllers with jk, including jobs, runs, logs, artifacts, credentials, nodes, queues, and plugins.
SkillAFreshness: A★ 3Claude Code +9 - 0.578score
sol
cskwork
Orchestrate application work through GPT-6.1 Sol subagents (high reasoning, fast tier).
SkillAFreshness: A★ 3Claude Code +9 - 0.578score
update-all
cskwork
Update all coding agent harnesses (Codex, Claude Code, pi, Antigravity CLI) in one run with before/after version reporting.
SkillAFreshness: A★ 3Claude Code +9 - 0.575score
call-agent
cskwork
Delegate a bounded task to a user-named peer CLI, or to fill a capability the host lacks.
SkillAFreshness: A★ 3Claude Code +9
Signals
- Stars
- 3
- Forks
- 1
- Open issues
- 0
- Last commit (this folder)
- Oct 9, 2026 (20 days ago)
- Last release
- Oct 24, 2026 (5 days ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/cskwork/skills
- Website
- cskwork.github.io/skills/
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from cskwork
- 0.586score
diagram-design
cskwork
Create branded architecture, architecture delta, IT current-state, flowchart, sequence, state machine, ER/data model, timeline, swimlane, quadrant…
SkillAFreshness: A★ 3Claude Code +9 - 0.586score
gpt-image-2
cskwork
Generate images with ChatGPT's current image model (GPT Image 2 at the time of writing, and whatever model succeeds it) inside your agent, using your existing…
SkillAFreshness: A★ 3Claude Code +9 - 0.586score
image-first-ui
cskwork
Image-first interface creation.
SkillAFreshness: A★ 3Claude Code +9 - 0.586score
motion-graphics
cskwork
Make code-driven motion-graphics videos rendered from HTML, frame-accurate and reproducible.
SkillAFreshness: A★ 3Claude Code +9 - 0.586score
bughunt
cskwork
Find and fix interaction bugs in games, CRUD apps, AI chat and media apps with seeded self-playing personas, per-step invariant oracles, reproducible failure…
SkillAFreshness: A★ 3Claude Code +9 - 0.586score
db-intelligence
cskwork
Unified database intelligence for PostgreSQL, MySQL, SQLite, and MongoDB in one skill: detect the engine, connect credential-safe, extract schema, build an…
SkillAFreshness: A★ 3Claude Code +9
Cite this page
mobile-ui-ux skill. Agent Mart. https://agentmart.dev/skills/cskwork-mobile-ui-ux (last updated 2026-10-24)