fix(skills): Add Taste Skill limitations

Add limitation sections to the Taste Skill import and plugin mirrors so the release audit no longer flags the new frontend design skills.

plugins/antigravity-awesome-skills-claude/skills/design-taste-frontend/SKILL.md

@@ -19,6 +19,12 @@ tools: [claude, cursor, codex, antigravity]
 - Use when React, Next.js, Tailwind, motion, component states, typography, spacing, color, or responsive behavior need senior-level design judgment.
 - Use when the output must override common LLM UI biases such as centered heroes, purple gradients, card overuse, poor states, and fragile layouts.
 
+## Limitations
+
+- This skill provides frontend design and implementation guidance; it does not replace project-specific product requirements, accessibility review, or user testing.
+- Verify framework versions, installed dependencies, responsive behavior, and build output in the target repository before treating generated UI as production-ready.
+- Do not force these design rules when the existing product, brand system, or platform conventions require a different visual direction.
+
 
 ## 1. ACTIVE BASELINE CONFIGURATION
 * DESIGN_VARIANCE: 8 (1=Perfect Symmetry, 10=Artsy Chaos)

plugins/antigravity-awesome-skills-claude/skills/full-output-enforcement/SKILL.md

@@ -19,6 +19,12 @@ tools: [claude, cursor, codex, antigravity]
 - Use when placeholder code, skipped sections, TODO stubs, or descriptions in place of implementation would break the request.
 - Use when a long answer may need clean continuation chunks without losing completeness or structural integrity.
 
+## Limitations
+
+- This skill enforces completeness, but it does not override token limits, safety constraints, missing source context, or user-provided scope boundaries.
+- Split long outputs into clearly labeled continuation chunks when necessary, and verify that each chunk connects cleanly to the previous one.
+- Do not invent unavailable code, credentials, private APIs, or project files to satisfy a request for complete output.
+
 
 ## Baseline
 

plugins/antigravity-awesome-skills-claude/skills/gpt-taste/SKILL.md

@@ -19,6 +19,12 @@ tools: [claude, cursor, codex, antigravity]
 - Use when GSAP, pinned scroll, scrubbing, card stacking, horizontal motion, or other advanced animation is appropriate.
 - Use when the output must avoid narrow six-line hero headings, cheap meta labels, empty bento cells, and generic left-right sections.
 
+## Limitations
+
+- This skill assumes a frontend project can support GSAP or equivalent animation libraries; check dependencies and performance budgets before implementation.
+- Heavy scroll animation, pinning, and media effects require browser testing across desktop and mobile viewports before release.
+- Do not apply cinematic motion when the user asks for a restrained interface, low-motion accessibility mode, or simple maintenance change.
+
 
 You are an elite, award-winning frontend design engineer. Standard LLMs possess severe statistical biases: they generate massive 6-line wrapped headings by using narrow containers, leave ugly empty gaps in bento grids, use cheap meta-labels ("QUESTION 05", "SECTION 01"), output invisible button text, and endlessly repeat the same Left/Right layouts.
 

plugins/antigravity-awesome-skills-claude/skills/high-end-visual-design/SKILL.md

@@ -19,6 +19,12 @@ tools: [claude, cursor, codex, antigravity]
 - Use when building a landing page, portfolio, SaaS UI, consumer product page, or app surface that needs premium depth and motion.
 - Use when the design must avoid generic fonts, harsh shadows, static layouts, default navbars, and ordinary Bootstrap-style grids.
 
+## Limitations
+
+- This skill is visual-design focused; it does not replace brand strategy, conversion research, accessibility validation, or production QA.
+- Premium fonts, icon sets, images, and motion libraries must exist in the target project or be added intentionally before generated code is used.
+- Avoid applying luxury motion and heavy visual treatments to constrained dashboards, regulated products, or low-performance environments.
+
 
 ## 1. Meta Information & Core Directive
 - **Persona:** `Vanguard_UI_Architect`

plugins/antigravity-awesome-skills-claude/skills/industrial-brutalist-ui/SKILL.md

@@ -19,6 +19,12 @@ tools: [claude, cursor, codex, antigravity]
 - Use when building data-heavy dashboards, portfolios, editorial pages, or command-center UIs that should feel raw and mechanical.
 - Use when a design must reject soft gradients, rounded consumer UI, glassmorphism, and generic SaaS card layouts.
 
+## Limitations
+
+- This style is intentionally severe and may not fit consumer products, accessibility-sensitive flows, or brands that require warmth and softness.
+- CRT, halftone, dithering, and degradation effects must be tested for readability, contrast, and motion sensitivity.
+- Do not mix the light industrial and dark telemetry palettes in the same interface unless the user explicitly asks for a controlled hybrid.
+
 
 ## 1. Skill Meta
 **Name:** Industrial Brutalism & Tactical Telemetry Interface Engineering

plugins/antigravity-awesome-skills-claude/skills/minimalist-ui/SKILL.md

@@ -19,6 +19,12 @@ tools: [claude, cursor, codex, antigravity]
 - Use when designing warm monochrome interfaces with crisp borders, generous whitespace, muted pastel accents, and quiet motion.
 - Use when the task should avoid gradients, heavy shadows, saturated colors, pill-heavy components, and generic SaaS visuals.
 
+## Limitations
+
+- Minimalism can hide hierarchy when content is dense; validate scannability, contrast, and navigation clarity with real content.
+- This skill assumes the product can support restrained palettes and typography-led layouts; do not override an established brand system without cause.
+- Subtle motion and flat surfaces still need responsive, keyboard, and screen-reader verification in the target project.
+
 
 ## 1. Protocol Overview
 Name: Premium Utilitarian Minimalism & Editorial UI

plugins/antigravity-awesome-skills-claude/skills/redesign-existing-projects/SKILL.md

@@ -19,6 +19,12 @@ tools: [claude, cursor, codex, antigravity]
 - Use when the task is to audit current frontend code and make targeted visual improvements without changing the product architecture.
 - Use when the design feels generic, AI-generated, poorly spaced, visually flat, or missing responsive, interactive, loading, empty, or error states.
 
+## Limitations
+
+- This skill upgrades existing UI but does not authorize framework migrations, information-architecture rewrites, or product-scope expansion by default.
+- Preserve working behavior, routing, data flows, accessibility semantics, and tests while making visual changes.
+- Validate redesigned screens in the actual app across supported browsers and viewport sizes before considering the work complete.
+
 
 ## How This Works
 

plugins/antigravity-awesome-skills-claude/skills/stitch-design-taste/SKILL.md

@@ -19,6 +19,12 @@ tools: [claude, cursor, codex, antigravity]
 - Use when translating premium frontend taste rules into Stitch-friendly visual descriptions, color roles, typography specs, and component behavior.
 - Use when the design system must prevent generic AI UI patterns before screens are generated.
 
+## Limitations
+
+- This skill produces semantic design-system guidance for Stitch; it does not guarantee Stitch will render every constraint exactly.
+- Generated `DESIGN.md` files still require review against the actual product brief, brand constraints, accessibility needs, and screen content.
+- Motion sections document implementation intent for later coding agents because Stitch itself may generate static screens.
+
 
 ## Overview
 This skill generates `DESIGN.md` files optimized for Google Stitch screen generation. It translates the battle-tested anti-slop frontend engineering directives into Stitch's native semantic design language — descriptive, natural-language rules paired with precise values that Stitch's AI agent can interpret to produce premium, non-generic interfaces.

plugins/antigravity-awesome-skills/skills/design-taste-frontend/SKILL.md

@@ -19,6 +19,12 @@ tools: [claude, cursor, codex, antigravity]
 - Use when React, Next.js, Tailwind, motion, component states, typography, spacing, color, or responsive behavior need senior-level design judgment.
 - Use when the output must override common LLM UI biases such as centered heroes, purple gradients, card overuse, poor states, and fragile layouts.
 
+## Limitations
+
+- This skill provides frontend design and implementation guidance; it does not replace project-specific product requirements, accessibility review, or user testing.
+- Verify framework versions, installed dependencies, responsive behavior, and build output in the target repository before treating generated UI as production-ready.
+- Do not force these design rules when the existing product, brand system, or platform conventions require a different visual direction.
+
 
 ## 1. ACTIVE BASELINE CONFIGURATION
 * DESIGN_VARIANCE: 8 (1=Perfect Symmetry, 10=Artsy Chaos)

plugins/antigravity-awesome-skills/skills/full-output-enforcement/SKILL.md

@@ -19,6 +19,12 @@ tools: [claude, cursor, codex, antigravity]
 - Use when placeholder code, skipped sections, TODO stubs, or descriptions in place of implementation would break the request.
 - Use when a long answer may need clean continuation chunks without losing completeness or structural integrity.
 
+## Limitations
+
+- This skill enforces completeness, but it does not override token limits, safety constraints, missing source context, or user-provided scope boundaries.
+- Split long outputs into clearly labeled continuation chunks when necessary, and verify that each chunk connects cleanly to the previous one.
+- Do not invent unavailable code, credentials, private APIs, or project files to satisfy a request for complete output.
+
 
 ## Baseline
 

plugins/antigravity-awesome-skills/skills/gpt-taste/SKILL.md

@@ -19,6 +19,12 @@ tools: [claude, cursor, codex, antigravity]
 - Use when GSAP, pinned scroll, scrubbing, card stacking, horizontal motion, or other advanced animation is appropriate.
 - Use when the output must avoid narrow six-line hero headings, cheap meta labels, empty bento cells, and generic left-right sections.
 
+## Limitations
+
+- This skill assumes a frontend project can support GSAP or equivalent animation libraries; check dependencies and performance budgets before implementation.
+- Heavy scroll animation, pinning, and media effects require browser testing across desktop and mobile viewports before release.
+- Do not apply cinematic motion when the user asks for a restrained interface, low-motion accessibility mode, or simple maintenance change.
+
 
 You are an elite, award-winning frontend design engineer. Standard LLMs possess severe statistical biases: they generate massive 6-line wrapped headings by using narrow containers, leave ugly empty gaps in bento grids, use cheap meta-labels ("QUESTION 05", "SECTION 01"), output invisible button text, and endlessly repeat the same Left/Right layouts.
 

plugins/antigravity-awesome-skills/skills/high-end-visual-design/SKILL.md

@@ -19,6 +19,12 @@ tools: [claude, cursor, codex, antigravity]
 - Use when building a landing page, portfolio, SaaS UI, consumer product page, or app surface that needs premium depth and motion.
 - Use when the design must avoid generic fonts, harsh shadows, static layouts, default navbars, and ordinary Bootstrap-style grids.
 
+## Limitations
+
+- This skill is visual-design focused; it does not replace brand strategy, conversion research, accessibility validation, or production QA.
+- Premium fonts, icon sets, images, and motion libraries must exist in the target project or be added intentionally before generated code is used.
+- Avoid applying luxury motion and heavy visual treatments to constrained dashboards, regulated products, or low-performance environments.
+
 
 ## 1. Meta Information & Core Directive
 - **Persona:** `Vanguard_UI_Architect`

plugins/antigravity-awesome-skills/skills/industrial-brutalist-ui/SKILL.md

@@ -19,6 +19,12 @@ tools: [claude, cursor, codex, antigravity]
 - Use when building data-heavy dashboards, portfolios, editorial pages, or command-center UIs that should feel raw and mechanical.
 - Use when a design must reject soft gradients, rounded consumer UI, glassmorphism, and generic SaaS card layouts.
 
+## Limitations
+
+- This style is intentionally severe and may not fit consumer products, accessibility-sensitive flows, or brands that require warmth and softness.
+- CRT, halftone, dithering, and degradation effects must be tested for readability, contrast, and motion sensitivity.
+- Do not mix the light industrial and dark telemetry palettes in the same interface unless the user explicitly asks for a controlled hybrid.
+
 
 ## 1. Skill Meta
 **Name:** Industrial Brutalism & Tactical Telemetry Interface Engineering

plugins/antigravity-awesome-skills/skills/minimalist-ui/SKILL.md

@@ -19,6 +19,12 @@ tools: [claude, cursor, codex, antigravity]
 - Use when designing warm monochrome interfaces with crisp borders, generous whitespace, muted pastel accents, and quiet motion.
 - Use when the task should avoid gradients, heavy shadows, saturated colors, pill-heavy components, and generic SaaS visuals.
 
+## Limitations
+
+- Minimalism can hide hierarchy when content is dense; validate scannability, contrast, and navigation clarity with real content.
+- This skill assumes the product can support restrained palettes and typography-led layouts; do not override an established brand system without cause.
+- Subtle motion and flat surfaces still need responsive, keyboard, and screen-reader verification in the target project.
+
 
 ## 1. Protocol Overview
 Name: Premium Utilitarian Minimalism & Editorial UI

plugins/antigravity-awesome-skills/skills/redesign-existing-projects/SKILL.md

@@ -19,6 +19,12 @@ tools: [claude, cursor, codex, antigravity]
 - Use when the task is to audit current frontend code and make targeted visual improvements without changing the product architecture.
 - Use when the design feels generic, AI-generated, poorly spaced, visually flat, or missing responsive, interactive, loading, empty, or error states.
 
+## Limitations
+
+- This skill upgrades existing UI but does not authorize framework migrations, information-architecture rewrites, or product-scope expansion by default.
+- Preserve working behavior, routing, data flows, accessibility semantics, and tests while making visual changes.
+- Validate redesigned screens in the actual app across supported browsers and viewport sizes before considering the work complete.
+
 
 ## How This Works
 

plugins/antigravity-awesome-skills/skills/stitch-design-taste/SKILL.md

@@ -19,6 +19,12 @@ tools: [claude, cursor, codex, antigravity]
 - Use when translating premium frontend taste rules into Stitch-friendly visual descriptions, color roles, typography specs, and component behavior.
 - Use when the design system must prevent generic AI UI patterns before screens are generated.
 
+## Limitations
+
+- This skill produces semantic design-system guidance for Stitch; it does not guarantee Stitch will render every constraint exactly.
+- Generated `DESIGN.md` files still require review against the actual product brief, brand constraints, accessibility needs, and screen content.
+- Motion sections document implementation intent for later coding agents because Stitch itself may generate static screens.
+
 
 ## Overview
 This skill generates `DESIGN.md` files optimized for Google Stitch screen generation. It translates the battle-tested anti-slop frontend engineering directives into Stitch's native semantic design language — descriptive, natural-language rules paired with precise values that Stitch's AI agent can interpret to produce premium, non-generic interfaces.

skills/design-taste-frontend/SKILL.md

@@ -19,6 +19,12 @@ tools: [claude, cursor, codex, antigravity]
 - Use when React, Next.js, Tailwind, motion, component states, typography, spacing, color, or responsive behavior need senior-level design judgment.
 - Use when the output must override common LLM UI biases such as centered heroes, purple gradients, card overuse, poor states, and fragile layouts.
 
+## Limitations
+
+- This skill provides frontend design and implementation guidance; it does not replace project-specific product requirements, accessibility review, or user testing.
+- Verify framework versions, installed dependencies, responsive behavior, and build output in the target repository before treating generated UI as production-ready.
+- Do not force these design rules when the existing product, brand system, or platform conventions require a different visual direction.
+
 
 ## 1. ACTIVE BASELINE CONFIGURATION
 * DESIGN_VARIANCE: 8 (1=Perfect Symmetry, 10=Artsy Chaos)

skills/full-output-enforcement/SKILL.md

@@ -19,6 +19,12 @@ tools: [claude, cursor, codex, antigravity]
 - Use when placeholder code, skipped sections, TODO stubs, or descriptions in place of implementation would break the request.
 - Use when a long answer may need clean continuation chunks without losing completeness or structural integrity.
 
+## Limitations
+
+- This skill enforces completeness, but it does not override token limits, safety constraints, missing source context, or user-provided scope boundaries.
+- Split long outputs into clearly labeled continuation chunks when necessary, and verify that each chunk connects cleanly to the previous one.
+- Do not invent unavailable code, credentials, private APIs, or project files to satisfy a request for complete output.
+
 
 ## Baseline
 

skills/gpt-taste/SKILL.md

@@ -19,6 +19,12 @@ tools: [claude, cursor, codex, antigravity]
 - Use when GSAP, pinned scroll, scrubbing, card stacking, horizontal motion, or other advanced animation is appropriate.
 - Use when the output must avoid narrow six-line hero headings, cheap meta labels, empty bento cells, and generic left-right sections.
 
+## Limitations
+
+- This skill assumes a frontend project can support GSAP or equivalent animation libraries; check dependencies and performance budgets before implementation.
+- Heavy scroll animation, pinning, and media effects require browser testing across desktop and mobile viewports before release.
+- Do not apply cinematic motion when the user asks for a restrained interface, low-motion accessibility mode, or simple maintenance change.
+
 
 You are an elite, award-winning frontend design engineer. Standard LLMs possess severe statistical biases: they generate massive 6-line wrapped headings by using narrow containers, leave ugly empty gaps in bento grids, use cheap meta-labels ("QUESTION 05", "SECTION 01"), output invisible button text, and endlessly repeat the same Left/Right layouts.
 

skills/high-end-visual-design/SKILL.md

@@ -19,6 +19,12 @@ tools: [claude, cursor, codex, antigravity]
 - Use when building a landing page, portfolio, SaaS UI, consumer product page, or app surface that needs premium depth and motion.
 - Use when the design must avoid generic fonts, harsh shadows, static layouts, default navbars, and ordinary Bootstrap-style grids.
 
+## Limitations
+
+- This skill is visual-design focused; it does not replace brand strategy, conversion research, accessibility validation, or production QA.
+- Premium fonts, icon sets, images, and motion libraries must exist in the target project or be added intentionally before generated code is used.
+- Avoid applying luxury motion and heavy visual treatments to constrained dashboards, regulated products, or low-performance environments.
+
 
 ## 1. Meta Information & Core Directive
 - **Persona:** `Vanguard_UI_Architect`

skills/industrial-brutalist-ui/SKILL.md

@@ -19,6 +19,12 @@ tools: [claude, cursor, codex, antigravity]
 - Use when building data-heavy dashboards, portfolios, editorial pages, or command-center UIs that should feel raw and mechanical.
 - Use when a design must reject soft gradients, rounded consumer UI, glassmorphism, and generic SaaS card layouts.
 
+## Limitations
+
+- This style is intentionally severe and may not fit consumer products, accessibility-sensitive flows, or brands that require warmth and softness.
+- CRT, halftone, dithering, and degradation effects must be tested for readability, contrast, and motion sensitivity.
+- Do not mix the light industrial and dark telemetry palettes in the same interface unless the user explicitly asks for a controlled hybrid.
+
 
 ## 1. Skill Meta
 **Name:** Industrial Brutalism & Tactical Telemetry Interface Engineering

skills/minimalist-ui/SKILL.md

@@ -19,6 +19,12 @@ tools: [claude, cursor, codex, antigravity]
 - Use when designing warm monochrome interfaces with crisp borders, generous whitespace, muted pastel accents, and quiet motion.
 - Use when the task should avoid gradients, heavy shadows, saturated colors, pill-heavy components, and generic SaaS visuals.
 
+## Limitations
+
+- Minimalism can hide hierarchy when content is dense; validate scannability, contrast, and navigation clarity with real content.
+- This skill assumes the product can support restrained palettes and typography-led layouts; do not override an established brand system without cause.
+- Subtle motion and flat surfaces still need responsive, keyboard, and screen-reader verification in the target project.
+
 
 ## 1. Protocol Overview
 Name: Premium Utilitarian Minimalism & Editorial UI

skills/redesign-existing-projects/SKILL.md

@@ -19,6 +19,12 @@ tools: [claude, cursor, codex, antigravity]
 - Use when the task is to audit current frontend code and make targeted visual improvements without changing the product architecture.
 - Use when the design feels generic, AI-generated, poorly spaced, visually flat, or missing responsive, interactive, loading, empty, or error states.
 
+## Limitations
+
+- This skill upgrades existing UI but does not authorize framework migrations, information-architecture rewrites, or product-scope expansion by default.
+- Preserve working behavior, routing, data flows, accessibility semantics, and tests while making visual changes.
+- Validate redesigned screens in the actual app across supported browsers and viewport sizes before considering the work complete.
+
 
 ## How This Works
 

skills/stitch-design-taste/SKILL.md

@@ -19,6 +19,12 @@ tools: [claude, cursor, codex, antigravity]
 - Use when translating premium frontend taste rules into Stitch-friendly visual descriptions, color roles, typography specs, and component behavior.
 - Use when the design system must prevent generic AI UI patterns before screens are generated.
 
+## Limitations
+
+- This skill produces semantic design-system guidance for Stitch; it does not guarantee Stitch will render every constraint exactly.
+- Generated `DESIGN.md` files still require review against the actual product brief, brand constraints, accessibility needs, and screen content.
+- Motion sections document implementation intent for later coding agents because Stitch itself may generate static screens.
+
 
 ## Overview
 This skill generates `DESIGN.md` files optimized for Google Stitch screen generation. It translates the battle-tested anti-slop frontend engineering directives into Stitch's native semantic design language — descriptive, natural-language rules paired with precise values that Stitch's AI agent can interpret to produce premium, non-generic interfaces.