Repository navigation
Releases
v1.2.0 — Lateral UX Thinking: Problem Diagnosis & High-Taste Fixes
✦ What's New in v1.2.0: Lateral UX Thinking & Root-Cause Fixes
Inspired by structured lateral thinking methodologies (such as Edward de Bono's lateral thinking and danium/lateral-thinking), this release equips AI assistants and human designers with an explicit thinking framework to diagnose problems and propose high-taste fixes without falling into predictable, one-size-fits-all AI clichés.
🚫 The 'Vanilla Trap' in AI UX Advice
Most AI agents reflexively prescribe the same superficial band-aids:
- The Tooltip Band-Aid
(i): Adding explanations instead of fixing confusing terminology or layouts. - The Modal/Banner Hammer: Forcing pop-ups over ignored buttons, causing banner blindness.
- The Bigger Button Fallacy: Enlarging CTAs when the user is actually terrified of being billed.
- The Generic Checklist Tour: Imposing multi-step tours before showing any value.
- The Fake Urgency Reflex: Slapping countdown timers on unconvincing offers.
🧠 The Lateral UX Thinking Protocol (references/10-problem-solving-lateral-ux.md)
-
Diagnose the 5 Friction Dimensions:
- Comprehension Friction: Mental model mismatch, insider jargon, weak information scent.
- Anxiety & Risk Friction: Loss aversion, fear of spam/billing/data loss, missing risk reversal.
- Cognitive & Motor Friction: Excessive reading, working memory overload, Fitts's distance.
- Timing & Value Asymmetry: Asking for commitment before demonstrating value ('Earn the ask').
- Autonomy & Reactance: Feeling coerced, nagged, or trapped (dark patterns).
-
The Abstraction Climb:
- Ask 'What is this a way of doing?' to detach from the current UI widget and unlock alternative mechanisms.
-
The 8 Lateral Thinking Moves:
- Subtractive Solving: Delete the step, infer the data, or eliminate the question via smart defaults.
- Timing Shift: Move the ask to the moment of post-value motivation (e.g. at export, not on landing).
- Effort Inversion: Shift labor from human fingers to software logic (Tesler's law).
- Contextual Relocation: Bring controls directly into inline focus/hover rather than remote toolbars.
- Radical Risk Reversal: Dissolve fear directly at the trigger ('No credit card required. Cancel in 1 click').
- State Reframing: Turn empty, loading, or 404 dead ends into high-momentum springboards.
- Progressive Engagement: Replace walls of effort with a 1-second starter commitment.
- Assumption Inversion: Challenge industry dogma and test the plausible opposite.
-
The Three-Angle Mandate & Disqualification Check:
- Propose 3 distinct structural directions (Angle A: Subtractive / System, Angle B: Cognitive / Clarity, Angle C: Structural / Timing Shift).
- Explicitly disqualify and reject superficial band-aids.
v1.1.0 — The 20 Core Taste Maxims & 19 Foundational UX Laws
✦ What's New in v1.1.0
This release integrates The 20 Core Taste Maxims and all 19 Foundational UX Laws into the skill architecture, complete with empirical evidence grades (A–D), psychological origins, and audit checklist criteria.
🎯 The Twenty Taste Maxims & UX Laws
- Reduce choices per screen — Hick's Law (Grade A/B,
01) - Make targets large — Fitts's Law (Grade A,
02) - Follow familiar patterns — Jakob's Law (Grade B,
01) - Group-related information — Law of Proximity (Grade A,
02) - Break content into chunks — Miller's Law & Chunking (Grade A,
01) - Interactions within 400 milliseconds — Doherty Threshold (Grade B/C,
05) - Highlight the primary action — Von Restorff Effect (Grade A/B,
02) - Place key actions nearby — Minimize Target Distance (Grade A,
02) - Put essentials first — Serial Position Effect & Pareto Principle (Grade A/B,
01,02) - End flows memorably — Peak-End Rule (Grade A,
05) - Show visible progress — Goal-Gradient Effect & Endowed Progress (Grade B,
04) - Simplify complex interfaces — Law of Prägnanz & Occam's Razor (Grade A/B,
01,02) - Use sensible defaults — Default Effect & Status Quo Bias (Grade A,
01,03) - Prevent errors proactively — Error Prevention (Poka-Yoke) (Grade B,
06) - Make errors recoverable — Postel's Law & Forgiveness (Grade B,
06) - Maintain pattern consistency — Consistency & Standards / Jakob's Law (Grade B,
01,06) - Connect related elements visually — Law of Uniform Connectedness & Similarity (Grade A,
02) - Reduce task completion time — Parkinson's Law (Grade B,
04) - Reveal complexity gradually — Progressive Disclosure & Tesler's Law (Grade B,
01) - Make completion feel closer — Endowed Progress & Ovsiankina Resumption (Grade B,
04)
📚 Module Enhancements
01-cognitive-load.md: Added Occam's razor (law of parsimony) and Pareto principle (80/20 rule).02-perception-attention-visual.md: Deepened Gestalt laws (Proximity, Similarity, Uniform Connectedness, Prägnanz) and Fitts's target distance minimization.04-motivation-habit-progress.md: Added Parkinson's law (task duration compression vs expansion).05-emotion-memory-delight.md: Expanded Doherty threshold (sub-400ms flow and sub-100ms feedback).08-evidence-myths-measurement.md: Updated the comprehensive Replication Status Table with all 19 laws.09-audit-checklist.md: Added corresponding audit checklist questions.
v1.0.0 — Initial Release: Research-Backed UX/UI Taste & Audit Skill
✦ UX/UI Taste v1.0.0 — Initial Release
Welcome to the initial release of UX/UI Taste, a research-backed library of cognitive principles, behavioral economics, and actionable heuristics for designing and auditing digital experiences that convert, satisfy, and feel good to use.
🧠 Core Foundations
- The 10 Taste Rules: Foundational defaults shaping every design and audit recommendation.
- Evidence Grading Framework (A–D): Clear distinction between replicated, robust empirical science and common design myths.
- Strict Anti-Dark-Pattern Standard: Grounded in the 3 ethical tests (Transparency, Regret, Symmetry) and aligned with US FTC, EU DSA (Art. 25), and UK DMCC regulations.
📚 9 Reference Modules
- 01. Cognitive Load: Working memory, Hick's law, progressive disclosure, mental models, plain language.
- 02. Perception & Visual Design: Visual hierarchy, Gestalt laws, scanning patterns, Fitts's law, contrast, typography.
- 03. Decision, Persuasion & Pricing: Defaults, anchoring, decoy effect, framing, loss aversion, price transparency.
- 04. Motivation & Progress: Fogg model, time-to-value, goal gradient, endowed progress, IKEA effect, habits.
- 05. Emotion, Memory & Delight: Peak-end rule, Kano model, perceived performance, labor illusion, error tone.
- 06. Feedback, Control & Forms: Heuristics, system status, form design, inline validation, checkout UX.
- 07. Ethics, Dark Patterns & Law: Dark pattern taxonomy, fair alternatives, US/EU/UK legal frameworks.
- 08. Evidence, Myths & Measurement: Debunked UX myths, effect size calibration, A/B testing hygiene.
- 09. Audit Checklist & Template: Comprehensive checklist with a 0–4 severity scale and ready-to-use audit report template.
🚀 Integration & Tools
- AI Agent Skill: Packaged as
SKILL.mdcompatible with Antigravity, Claude Code, Claude Desktop, Cursor, and custom agent systems. - Community & Standards: Full contribution guide (
CONTRIBUTING.md), Code of Conduct, GitHub Issue and PR templates, and automated CI validation workflow.