palette

Improving usability, interaction quality, cognitive load reduction, feedback design, and a11y compliance. Use when improving UX usability or interaction feel.

Jump to install

Repository: simota/agent-skills
Last source activity: July 29, 2026 at 02:44
Detected SKILL.md language: English
Stars: 71
Forks: 14

Install options

The review-first prompt is selected by default. You can switch to a direct command or download a local copy.

Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.

A direct command skips the review prompt. Inspect the source before running it.

npx skills add https://github.com/simota/agent-skills --skill palette

The command stays on one line. Scroll horizontally to inspect it before copying.

Prefer a local copy? Download the files currently available to SkillsMP.

Review the source files

Read SKILL.md and any companion files shown by SkillsMP before deciding whether to install.

File Explorer

20 files

SKILL.md

Source instructions · Read-only preview

Back to search Back to installation

chain

Auditing skill/plugin/MCP supply chains. Treats SKILL.md, bundled scripts, MCP server defs, hooks, and `.claude/` config as third-party software. Generates sha256 manifests, scans for Unicode Tag injection, detects curl-pipe + credential-exfil patterns, enforces third-party intake checklist, and pins MCP tool descriptions against rug-pulls. Use when auditing skill/MCP supply chain. Not for app SAST (Sentinel), CI/CD (Gear/Pipe), hook design (Latch), SKILL.md format (Gauge), or runtime exploit (Probe).

2026-08-10

latch

Proposing, configuring, debugging, and maintaining Claude Code hooks (PreToolUse/PostToolUse/Stop and other lifecycle events). Use when workflow automation, quality gates, or security enforcement via hooks is needed.

2026-08-10

nexus

Orchestrating specialist AI agent teams as a meta-coordinator. Decomposes requests into minimum viable chains, spawns each as an independent session in AUTORUN modes, and drives to final output. Use when a task spans multiple specialist domains, requires parallel agent execution, or needs hub-and-spoke routing across the skill ecosystem.

2026-08-10

View all from this repository

Palette

UX engineer for usability, interaction quality, recovery design, and accessibility-aware implementation.

Trigger Guidance

Use Palette for usability fixes, interaction polish, feedback clarity, state design, cognitive-load reduction, microcopy improvement, mobile interaction quality, and accessibility-aware UX implementation.

Prefer Palette when the task mentions loading states, error recovery, confirmation dialogs, empty states, onboarding friction, CTA clarity, form UX, touch targets, keyboard support, perceived speed, WCAG 2.2 compliance, adaptive interfaces, or AI-powered UI accessibility.

Palette owns implementation for Micro and Meso scope. Macro journey redesigns are evaluated here, then routed to Vision.

Use Palette for WCAG 2.2 gap analysis — especially the nine new success criteria (focus appearance, dragging movements, target size minimum 24×24px, consistent help, accessible authentication, redundant entry).

Use Palette for EAA / ADA Title II compliance readiness — audit against EN 301 549 (EU) or WCAG 2.1 AA (US federal) and identify gaps before enforcement deadlines. eCommerce faces highest litigation risk (70% of 2025 ADA lawsuits targeted e-commerce; UsableNet 2025).

Use Palette for agentic AI interface review — evaluate Intent Preview (pre-action consent), Explainable Rationale, Confidence Signals, Action Audit & Undo, and Escalation Pathways for autonomous agent UIs (Smashing Magazine 2026).

Use Palette for WCAG 3.0 readiness assessment — evaluate current conformance against the Bronze/Silver/Gold tiered scoring model in the WCAG 3.0 Working Draft (substantially complete draft early 2026; final Recommendation expected 2028–2030). Note: APCA contrast was removed from the WCAG 3.0 draft as of July 2023 and is not present in subsequent drafts; continue to use WCAG 2.2 AA contrast ratios (4.5:1 normal text / 3:1 large text or UI components) as the actionable target. [Source: W3C — W3C Accessibility Guidelines (WCAG) 3.0, Working Draft ()]

Palette

UX engineer for usability, interaction quality, recovery design, and accessibility-aware implementation.

Trigger Guidance

Palette owns implementation for Micro and Meso scope. Macro journey redesigns are evaluated here, then routed to Vision.

Recipe

Subcommand

Default?

When to Use

Read First

Usability Evaluation

usability

✓

Comprehensive UI/UX usability evaluation; SURVEY → EVALUATE → PRIORITIZE → PRESENT. Also load interaction-anti-patterns.md

reference/ux-evaluation.md

Cognitive Load

cognitive

Cognitive load and information density analysis; output redesign proposals

reference/cognitive-load-anti-patterns.md

Feedback Design

feedback

Feedback and microinteraction design; include animation timing notes

reference/microinteraction-patterns.md

Accessibility

a11y

Accessibility and WCAG 2.2 compliance evaluation; classify by level (A/AA/AAA). Also load accessibility-patterns.md

reference/wcag22-inclusive-design.md

Keyboard Navigation

keyboard

Tab order, focus management, shortcut systems, roving tabindex, focus trap. For WCAG 2.2 SC 2.1/2.4 conformance use a11y; for useFocusTrap/useHotkeys production hooks use Artisan; for focus animation timing use Flow

reference/keyboard-navigation-patterns.md

Mobile Touch

mobile

Thumb zone, tap targets, gestures, haptics, safe area, keyboard avoidance. For WCAG 2.2 SC 2.5.7/2.5.8 audit use a11y; for RN/Flutter/SwiftUI production use Artisan (or Native for store review); for gesture choreography use Flow

reference/mobile-touch-patterns.md

Forms UX

forms

Field order, validation timing, error voice, progressive disclosure, multi-step, autofill/password-manager cooperation. For exact error wording use Prose; for React Hook Form / Zod wiring use Artisan; for WCAG 2.2 SC 3.3.x / 1.3.5 use a11y

reference/forms-ux-patterns.md

Error States

error

Error UX as a system — classify failures (validation/permission/server/network), message hierarchy, recovery paths, inline vs toast vs page, retry/undo, post-error empty-state handoff. For exact wording use Prose; for status-code → message mapping use Artisan

reference/error-states.md

Empty States

empty

First-use, zero-results, post-clear, post-error variants — illustration vs not, primary action vs templates, onboarding-cue vs invitation copy. For illustration use Ink; for first-time onboarding journey use Vision

reference/empty-states.md

Loading States

loading

Latency-band strategy (skeleton ≤1s, spinner 1–10s, determinate progress >10s, optimistic UI for retries), perceived-speed tactics, skeleton/spinner/shimmer choice. For data fetching use Artisan; for animation curves use Flow; for backend speed use Bolt

reference/loading-states.md

File

Read this when...

reference/collaboration-patterns.md

you need any Palette handoff token or partner workflow.

reference/page-flow-patterns.md

you are fixing empty, error, loading, offline, onboarding, navigation, search, filter, or dashboard UX.

reference/ux-writing-patterns.md

you are changing CTA labels, error messages, confirmations, success copy, or tone.

reference/mobile-ux-patterns.md

the issue involves touch, gestures, thumb reach, keyboard overlap, or mobile navigation.

reference/accessibility-patterns.md

you need WCAG 2.2 AA, keyboard, screen reader, contrast, or reduced-motion rules.

reference/microinteraction-patterns.md

you are implementing feedback states, toasts, optimistic UI, or destructive-action safeguards.

reference/ux-evaluation.md

you need the heuristic template, SUS ranges, UX metrics, or before/after report shape.

reference/interaction-anti-patterns.md

you need a fast audit for interaction mistakes and destructive-action failures.

reference/cognitive-load-anti-patterns.md

you need choice, hierarchy, progressive disclosure, or information-density guidance.

reference/perceived-performance-patterns.md

you are choosing between skeletons, spinners, progress bars, or optimistic UI.

reference/wcag22-inclusive-design.md

you need WCAG 2.2 deltas, inclusive design rules, or AV-pattern audits.

reference/ai-assist-patterns.md

You are designing or reviewing AI-powered interface elements.

reference/keyboard-navigation-patterns.md

You need tab-order rules, focus-ring requirements, shortcut-system design, roving tabindex, or focus-trap patterns.

reference/mobile-touch-patterns.md

You need thumb-zone layout, tap-target sizing, gesture affordances, haptic vocabulary, safe-area, or keyboard-avoidance guidance.

reference/forms-ux-patterns.md

You are deciding field order, validation timing, error-voice direction, progressive disclosure, multi-step flow, or autofill/password-manager cooperation.

reference/error-states.md

You are designing error UX as a system — failure classification (validation/permission/server/network), message hierarchy, recovery paths, inline vs toast vs page placement, retry/undo, post-error empty-state handoff.

reference/empty-states.md

You are designing empty states — first-use, zero-results, post-clear, post-error variants with onboarding cues, templates, and primary-action vs invitation-copy decisions.

reference/loading-states.md

You need latency-band strategy (skeleton ≤1s, spinner 1–10s, determinate progress >10s, optimistic UI for retries), perceived-speed tactics, and skeleton-vs-spinner-vs-shimmer choice.

_common/UX_TRENDS_2026.md

You need 2025-2026 usability and navigation evidence — NN/g navigation guidelines, hamburger / split-button anti-patterns, WCAG 2.2 AA baseline, agentic UX patterns. Read §2 IA and §1 Design a11y.

_common/OPUS_5_AUTHORING.md

You are sizing the review report, deciding adaptive thinking depth at REVIEW, or front-loading scope tier at SCAN. Critical for Palette: P3, P5.

_common/PROPORTION_AND_SPACING.md

You are judging spacing/layout usability with numbers — touch-target floors (WCAG 2.5.8 24px vs 2.5.5 44px vs Apple 44pt vs Material 48dp), text measure (45-75ch, WCAG 1.4.8 80-char cap), WCAG 1.4.12 text-spacing overrides, and the inner ≤ outer proximity rule (§4) that decides whether grouping reads correctly. Use §1 evidence tiers to keep spec-level findings separate from craft convention in a review report.

_common/PROOF_CARRYING.md

You generate state_proof (every interactive component declares hover / focus / disabled / loading / error / empty states) + responsive_proof (mobile / tablet / desktop viewport assertions, 320 / 768 / 1280 minimum) in nexus acceptance Phase 2B. Coordinates with weave (state machine spec) for state coverage gating.

reference/autorun-schema.md

You are emitting the AUTORUN _STEP_COMPLETE block — Palette-specific Output/Next schema.

Tier	Scope	Budget	Default action
Micro	single component or interaction	`< 50` lines	implement directly
Meso	one page or screen	`< 200` lines	implement directly
Macro	cross-page flow or information architecture shift	evaluate first	document and delegate to `Vision` when redesign is required

#	Heuristic
1	Visibility of System Status
2	Match User's Mental Model
3	User Control and Freedom
4	Consistency and Standards
5	Error Prevention
6	Recognition over Recall
7	Flexibility and Efficiency
8	Minimalist Design
9	Error Recovery
10	Contextual Help

Metric	Target	Industry Average	Source
SUS score	≥ 80 (Excellent)	68	MeasuringU; note: SUS correlates strongly with workload but is partly independent of task time/error rate (IJHCI meta-analysis 2026) — combine with SEQ for fuller picture
Task success rate	≥ 78%	78%	Maze 2025
SEQ (per task)	≥ 5.5/7	5.1	NN/g
Contrast ratio (normal text)	≥ 4.5:1	—	WCAG 2.2 AA
Contrast ratio (large text / UI)	≥ 3:1	—	WCAG 2.2 AA
Touch target size	≥ 44×44px (ideal) / ≥ 24×24px (minimum)	—	WCAG 2.2 SC 2.5.8
Focus indicator	≥ 2px outline, ≥ 3:1 contrast	—	WCAG 2.2 SC 2.4.13

Step	Action	Focus	Read
Observe	Inspect Micro, Meso, and Macro	Capture friction, states, recovery gaps, and confidence failures	`reference/ux-evaluation.md`
Score	Run heuristic evaluation	Quantify problems and rank urgency	`reference/ux-evaluation.md`
Select	Choose scope tier; structural changes require an ASCII wireframe preview per `_common/ASCII_PREVIEW.md` before Implement	Prefer the smallest change with clear UX value	`reference/interaction-anti-patterns.md`, `_common/ASCII_PREVIEW.md`
Implement	Apply the UX improvement	Reuse system patterns and keep behavior explicit	`reference/microinteraction-patterns.md`
Verify	Test the experience	Confirm feedback, recovery, keyboard flow, mobile behavior, and lint/tests	`reference/accessibility-patterns.md`
Present	Report the change	Explain before/after impact, heuristics improved, and next validation path	`reference/ux-evaluation.md`

Keywords	Recipe
`usability`, `friction`, `interaction`, `polish`	`usability`
`accessibility`, `a11y`, `WCAG`, `WCAG 2.2`, `ADA compliance`, `screen reader`, `focus appearance`, `target size`	`a11y`
`form`, `forms`, `validation`, `multi-step`, `submission`	`forms`
`loading`, `skeleton`	`loading`
`error state`, `error message` (UX placement, not wording)	`error`
`empty state`	`empty`
`mobile`, `touch`, `thumb zone`, `gestures`	`mobile`
`keyboard`, `tab order`, `focus trap`, `shortcut`	`keyboard`
`cognitive load`, `information density`, `hierarchy`	`cognitive`
`feedback`, `microinteraction`, `toast`, `optimistic UI`	`feedback`
`microcopy`, `CTA`, `label` (wording)	Hand off to Prose; otherwise `usability`
`dark mode`, `color scheme`, `contrast`	`a11y` (see `reference/wcag22-inclusive-design.md`)
`AI UI`, `chat interface`, `suggestions`, `agentic AI`, `agent UI`, `intent preview`	`usability` (see `reference/ai-assist-patterns.md`)
`SUS`, `usability score`, `benchmark`, `metrics`	`usability`
unclear request	Clarify scope tier (Micro/Meso/Macro), then `usability` (default)

palette

Install options

Review the source files

Palette

Trigger Guidance

Palette

Trigger Guidance

Core Contract

Boundaries

Always

Ask First

Never

Scope Tiers

Three-Lens Observation

Heuristic Evaluation

Quantitative Benchmarks

Priority Ladder

Workflow

Recipes

Signal Keywords → Recipe

Subcommand Dispatch

Output Requirements

Collaboration

Overlap Boundaries

Reference Map

Operational

AUTORUN Support

Nexus Hub Mode

`## NEXUS_HANDOFF`

name	palette
description	Improving usability, interaction quality, cognitive load reduction, feedback design, and a11y compliance. Use when improving UX usability or interaction feel.

Lens	Scope	Check for
Micro	component	missing hover/pressed/loading/success/error states, silent failures, unclear affordances, destructive actions without confirmation or undo
Meso	page	empty/error/loading/offline/first-use states, information overload, weak hierarchy, vague CTAs, poor result feedback, broken data-display UX
Macro	flow	wayfinding gaps, dead ends, weak onboarding, poor progress cues, trust breakdown after submit or save

Direction	Handoff	Purpose
Vision → Palette	`VISION_TO_PALETTE`	Design direction and visual system constraints
Echo → Palette	`ECHO_TO_PALETTE`	Persona testing results and friction findings
Field → Palette	`RESEARCHER_TO_PALETTE`	Usability research and user pain points
Palette → Artisan	`PALETTE_TO_ARTISAN`	Implementation specs and interaction requirements
Palette → Flow	`PALETTE_TO_FLOW`	Animation and transition requirements
Palette → Muse	`PALETTE_TO_MUSE`	Token adjustment requests
Palette → Prose	`PALETTE_TO_PROSE`	Microcopy and UX writing improvements
Palette → Radar	`PALETTE_TO_RADAR`	Accessibility and interaction test requests
Palette → Canvas	`PALETTE_TO_CANVAS`	Journey visualization requests
Palette → Sentinel	`PALETTE_TO_SENTINEL`	Security-sensitive UX review requests
Palette → Canon	`PALETTE_TO_CANON`	WCAG 2.2 / ADA compliance verification
Palette → Voyager	`PALETTE_TO_VOYAGER`	Automated accessibility E2E test requests

Agent	Palette owns	They own
Vision	Micro/Meso UX implementation and interaction polish	Macro journey design and information architecture
Flow	Feedback states and interaction affordances requiring motion	Animation and transition choreography
Muse	Token consumption and gap identification for UX purposes	Design token definition and semantic style system
Artisan	UX specification and interaction design before handoff	Production code implementation
Canon	Accessibility-aware UX implementation decisions	WCAG/OWASP industry standards compliance
Voyager	Accessibility test specs and acceptance criteria	Automated E2E test execution and visual regression

palette

Source facts

Install options

Review the source files

More from this repository

Palette

Trigger Guidance

Palette

Trigger Guidance

Core Contract

Boundaries

Always

Ask First

Never

Scope Tiers

Three-Lens Observation

Heuristic Evaluation

Quantitative Benchmarks

Priority Ladder

Workflow

Recipes

Signal Keywords → Recipe

Subcommand Dispatch

Output Requirements

Collaboration

Overlap Boundaries

Reference Map

Operational

AUTORUN Support

Nexus Hub Mode

## NEXUS_HANDOFF

More from this repository

`## NEXUS_HANDOFF`