Source profileQuality 82/100

nexu-io/open-design/skills/design-brief/SKILL.md

design-brief

Parse a structured design brief written in I-Lang protocol format into a concrete design spec. Eliminates ambiguity from vague requests like "make it professional" by requiring explicit dimensions: palette, typography, layout, mood, density, and constraints. Trigger keywords: "design brief", "create a design brief", "ilang brief", "structured brief".

Source repository stars
82,073
Declared platforms
0
Static risk flags
1
Last source update
2026-07-28
Source checked
2026-07-28

Decision brief

What it does—and where it fits

Parse a structured design brief into a concrete DESIGN.md and optional visual preview. Agent, follow this workflow exactly.

Best for

    Not for

    • Tasks that require unconfirmed production actions or broad system permissions.
    • Environments where the pinned source and install steps cannot be inspected.

    Compatibility matrix

    Platform support, with evidence labels

    PlatformStatusEvidenceWhat to check
    CodexNot declaredNo explicit evidencePortability before use
    Claude CodeNot declaredNo explicit evidencePortability before use
    CursorNot declaredNo explicit evidencePortability before use
    Gemini CLINot declaredNo explicit evidencePortability before use
    Open the compatibility checker

    Installation

    Inspect first. Install second.

    The source command is displayed only when detected. A safe inspection prompt is always available so your agent can explain every action before execution.

    Source-detected install commandSource
    npx skills add https://github.com/nexu-io/open-design --skill "skills/design-brief"
    Safe inspection promptEditorial

    Inspect the Agent Skill "design-brief" from https://github.com/nexu-io/open-design/blob/89d6d4ef21baf80f871595abdf6f7de6e941dd44/skills/design-brief/SKILL.md at commit 89d6d4ef21baf80f871595abdf6f7de6e941dd44. List every install step, command, network request, credential, file read/write, external action, and rollback step. Explain whether it fits my task. Do not install or execute anything until I approve.

    Workflow

    What the source asks the agent to do

    1. 01

      Background

      The 8 dimensions in this skill were derived from Open Design's original 71-system catalog; 71 is the research sample, not the current bundled-system count. Every DESIGN.md in design-systems/ still resolves at minimum: color palette, accent, typography, display font, layout model…

      The 8 dimensions in this skill were derived from Open Design's original 71-system catalog; 71 is the research sample, not the current bundled-system count. Every DESIGN.md in design-systems/ still resolves at minimum: c…Dimensions intentionally excluded from the brief level: animation timing, responsive strategy, and accessibility contrast. These are enforced at the template level by individual skills (e.g., saas-landing handles its ow…
    2. 02

      1. Accept input

      The user provides a design brief in one of two formats:

      The user provides a design brief in one of two formats:"I need a landing page for a developer tool. Clean, minimal, dark mode. Inter font. No flashy animations."If the user provides Option B, convert it to the structured format using the mapping table below, then proceed. Identify every dimension explicitly stated and flag dimensions that were left unspecified.
    3. 03

      Option A: I-Lang structured brief

      Review the “Option A: I-Lang structured brief” section in the pinned source before continuing.

      Review and apply the “Option A: I-Lang structured brief” source section.
    4. 04

      Option B: Natural language

      "I need a landing page for a developer tool. Clean, minimal, dark mode. Inter font. No flashy animations."

      "I need a landing page for a developer tool. Clean, minimal, dark mode. Inter font. No flashy animations."If the user provides Option B, convert it to the structured format using the mapping table below, then proceed. Identify every dimension explicitly stated and flag dimensions that were left unspecified.
    5. 05

      Natural language → I-Lang mapping

      For each sentence in the natural language input, identify dimension keywords and map to the closest structured value:

      For each sentence in the natural language input, identify dimension keywords and map to the closest structured value:When a phrase maps to multiple dimensions (e.g. "clean dark landing page" → mood=professionalminimal + palette=monochromedark + layout=singlecolumn), resolve each dimension independently. When multiple values are listed…

    Permission review

    Static risk signals and limitations

    Writes files

    medium · line 202

    The documentation asks the agent to create, modify, or delete local files.

    Create a single HTML file that visually renders the resolved design tokens. The preview must contain these 4 sections in order:

    Evidence record

    Why each signal appears

    EvidenceSourceComputedTestedEditorial
    SignalValueEvidence typeMeaning
    Quality score82/100ComputedDocumentation, specificity, maintenance, and trust rules
    Repository stars82,073SourceRepository attention, not individual Skill quality
    Compatibility0 platformsSourceDeclared in the catalog source record
    Usage guideautomated source guideEditorialGenerated or reviewed according to the visible evidence level

    Pinned source

    Provenance and original SKILL.md

    Repository
    nexu-io/open-design
    Skill path
    skills/design-brief/SKILL.md
    Commit
    89d6d4ef21baf80f871595abdf6f7de6e941dd44
    License
    Apache-2.0
    Collected
    2026-07-28
    Default branch
    main
    View the original SKILL.md

    Design Brief Skill

    Parse a structured design brief into a concrete DESIGN.md and optional visual preview. Agent, follow this workflow exactly.

    Background

    The 8 dimensions in this skill were derived from Open Design's original 71-system catalog; 71 is the research sample, not the current bundled-system count. Every DESIGN.md in design-systems/ still resolves at minimum: color palette, accent, typography, display font, layout model, and component style. We distilled these into 8 orthogonal dimensions that cover the decisions a designer makes before any pixel is placed. Mood and density were added because they are the two most common sources of ambiguity in natural language briefs ("make it clean" means different things to different people).

    Dimensions intentionally excluded from the brief level: animation timing, responsive strategy, and accessibility contrast. These are enforced at the template level by individual skills (e.g., saas-landing handles its own responsive logic), though the generated DESIGN.md includes sensible breakpoint defaults for downstream consumption.

    1. Accept input

    The user provides a design brief in one of two formats:

    Option A: I-Lang structured brief

    [PLAN:@DESIGN|type=saas_landing]
      |palette=navy_and_white|accent=coral
      |typography=inter|display=space_grotesk
      |layout=single_column|max_width=1200px
      |mood=professional_minimal
      |density=spacious|section_gap=96px
      |hero=headline+subhead+cta
      |sections=features,pricing,testimonials,footer
      |exclude=animations,parallax,gradients
      |responsive=mobile_first
    

    Option B: Natural language

    "I need a landing page for a developer tool. Clean, minimal, dark mode. Inter font. No flashy animations."

    If the user provides Option B, convert it to the structured format using the mapping table below, then proceed. Identify every dimension explicitly stated and flag dimensions that were left unspecified.

    Natural language → I-Lang mapping

    For each sentence in the natural language input, identify dimension keywords and map to the closest structured value:

    Natural language phraseDimensionI-Lang value
    "dark mode", "dark theme"palettemonochrome_dark
    "light", "white background"palettelight_clean
    "earthy", "warm tones"paletteearth_tones
    "pop of color", "vibrant"accentelectric_blue (default) or coral
    "subtle accent"accentmuted_sage (default) or slate
    "clean", "minimal", "simple"moodprofessional_minimal
    "playful", "fun", "friendly"moodplayful
    "bold", "brutalist", "raw"moodbrutalist
    "editorial", "magazine-like"moodeditorial
    "spacious", "lots of whitespace"densityspacious
    "compact", "dense", "information-rich"densitycompact
    "Inter", "system font"typographyinter (default) or system_ui
    "serif", "traditional"typographygeorgia (default) or playfair
    "monospace", "code-like"typographyjetbrains_mono
    "no animations", "static"excludeanimations
    "no gradients"excludegradients
    "no stock photos"excludestock_photos
    "single page"layoutsingle_column
    "two columns", "sidebar"layouttwo_column
    "mobile first"responsivemobile_first

    When a phrase maps to multiple dimensions (e.g. "clean dark landing page" → mood=professional_minimal + palette=monochrome_dark + layout=single_column), resolve each dimension independently. When multiple values are listed for a single mapping, the first is the default; the agent may select the alternative only if surrounding context strongly favors it.

    2. Validate dimensions

    Every design brief must resolve these 8 dimensions. If any are missing from the input, select sensible defaults using the rules in Section 2.2.

    The values listed below form a closed vocabulary. Only values in this table have concrete token mappings in Section 2.1. If the user provides a value not listed here, the agent must prompt for clarification rather than guessing.

    #DimensionKeyExample values
    1Color palettepalettenavy_and_white, earth_tones, monochrome_dark, light_clean
    2Accent coloraccentcoral, electric_blue, emerald, muted_sage
    3Body typographytypographyinter, system_ui, dm_sans, georgia
    4Display typographydisplayspace_grotesk, clash_display, same_as_body, playfair
    5Layout modellayoutsingle_column, two_column, asymmetric
    6Moodmoodprofessional_minimal, playful, brutalist, editorial
    7Densitydensitycompact, balanced, spacious
    8Constraintsexcludeanimations, gradients, stock_photos, carousel

    2.1 Symbolic → concrete token resolution

    Each symbolic value maps to concrete design tokens. The agent must resolve these before writing DESIGN.md:

    Symbolic valueConcrete tokens
    palette=navy_and_whiteBackground: #0F172A, Surface: #1E293B, Text: #F8FAFC, Secondary: #94A3B8
    palette=monochrome_darkBackground: #09090B, Surface: #18181B, Text: #FAFAFA, Secondary: #A1A1AA
    palette=light_cleanBackground: #FFFFFF, Surface: #F8FAFC, Text: #0F172A, Secondary: #64748B
    palette=earth_tonesBackground: #FFFBEB, Surface: #FEF3C7, Text: #451A03, Secondary: #92400E
    accent=coralAccent: #F97316, Hover: #EA580C
    accent=electric_blueAccent: #3B82F6, Hover: #2563EB
    accent=emeraldAccent: #10B981, Hover: #059669
    accent=muted_sageAccent: #84A98C, Hover: #6B8F73
    accent=slateAccent: #64748B, Hover: #475569
    typography=interBody: Inter, 400, 1rem/1.6
    typography=system_uiBody: system-ui, 400, 1rem/1.6
    typography=dm_sansBody: DM Sans, 400, 1rem/1.6
    typography=georgiaBody: Georgia, 400, 1.125rem/1.7
    display=space_groteskDisplay: Space Grotesk, 700, clamp(2rem, 5vw, 3.5rem)
    display=clash_displayDisplay: Clash Display, 700, clamp(2rem, 5vw, 3.5rem)
    display=playfairDisplay: Playfair Display, 700, clamp(2rem, 5vw, 3.5rem)
    display=same_as_bodyDisplay inherits body font family, weight 600
    density=compactSection spacing: 48px, Content padding: 16px/24px
    density=balancedSection spacing: 72px, Content padding: 24px/40px
    density=spaciousSection spacing: 96px, Content padding: 24px/48px

    Symbolic values not in this table are not valid. If the user provides an unrecognized value (e.g., palette=ocean_blue), the agent must prompt for clarification: "I don't recognize palette=ocean_blue. Did you mean navy_and_white, monochrome_dark, light_clean, or earth_tones?"

    2.2 Default resolution rules

    When a dimension is unspecified, defaults are selected based on mood compatibility:

    Unspecified dimensionDefault rule
    paletteIf mood=editorial → light_clean. If mood=brutalist → monochrome_dark. Otherwise → light_clean.
    accentIf palette is dark → coral. If palette is light → electric_blue.
    typographyAlways → inter (highest cross-platform legibility).
    displayIf mood=editorial → playfair. If mood=brutalist → space_grotesk. Otherwise → same_as_body.
    layoutAlways → single_column (safest responsive default).
    moodAlways → professional_minimal (least opinionated).
    densityAlways → balanced.
    excludeAlways → none (no constraints unless specified).

    If mood is also unspecified, all defaults fall back to the safe neutral set: palette=light_clean, accent=electric_blue, typography=inter, display=same_as_body, layout=single_column, mood=professional_minimal, density=balanced, exclude=none.

    3. Generate DESIGN.md

    This skill generates a new DESIGN.md from scratch based on the resolved brief dimensions. If a DESIGN.md already exists in the working directory, the agent should ask the user whether to overwrite or skip.

    Produce a DESIGN.md using the nine-section outline below. This outline is the skill's portable standalone output, inherited from Open Design's original upstream baseline; it is not the current repository package schema. Current bundled packages also carry manifest.json, tokens.css, and optional rich resources, while legacy and user-installed DESIGN.md-only content remains readable. All color hex values, font stacks, and spacing values must come from the resolved tokens in Section 2.1 — do not invent values outside the resolution table.

    # [Project Name] Design System
    
    ## Visual Theme & Atmosphere
    - Mood: [resolved from mood]
    - Feel: [derived from mood — e.g., professional_minimal → "Clean, confident, restrained"]
    - References: [if mood=editorial → "Magazine layouts, Monocle, Cereal"; if mood=brutalist → "Exposed structure, raw typography"]
    
    ## Color Palette & Roles
    - Background: [resolved from palette]
    - Surface: [resolved from palette]
    - Text primary: [resolved from palette]
    - Text secondary: [resolved from palette]
    - Accent: [resolved from accent]
    - Accent hover: [resolved from accent]
    
    ## Typography Rules
    - Display: [resolved from display], 700, clamp(2rem, 5vw, 3.5rem)
    - Body: [resolved from typography], 400, 1rem/1.6
    - Mono: JetBrains Mono, 400, 0.875rem
    
    ## Component Stylings
    - Buttons: [if mood=playful → "rounded-full", otherwise → "rounded-md"], accent bg, contrast text
    - Cards: surface bg, subtle border, 12px radius
    - Inputs: [if mood=brutalist → "thick border", otherwise → "transparent bg, bottom border"]
    
    ## Layout Principles
    - Max width: 1200px
    - Grid: [resolved from layout]
    - Section spacing: [resolved from density]
    - Content padding: [resolved from density]
    
    ## Depth & Elevation
    - Shadows: [if mood=brutalist → "hard 4px offset", if mood=professional_minimal → "none", otherwise → "subtle sm"]
    - Borders: 1px solid [derived from palette, 8% opacity of text color]
    
    ## Do's and Don'ts
    - DO use the declared color tokens exclusively.
    - DO maintain consistent section spacing.
    - DO ensure all text meets WCAG AA contrast ratio.
    - DON'T invent colors outside the palette.
    - DON'T add decorative shadows unless Depth & Elevation allows them.
    - DON'T use more than 2 display/body typefaces (monospace is a utility face for code and data — it does not count toward this limit).
    
    ## Responsive Behavior
    - Breakpoints: 640px (sm), 768px (md), 1024px (lg), 1280px (xl)
    - Mobile: single column, stack all sections vertically
    - Tablet: allow 2-column feature grids
    - Desktop: full layout with max-width constraint
    - Images: fluid, max-width 100%, maintain aspect ratio
    
    ## Agent Prompt Guide
    - Do NOT invent colors outside this palette.
    - Do NOT add box-shadows unless specified above.
    - Accent color appears maximum 3 times per viewport.
    - All interactive elements need :focus-visible outline.
    - [if exclude contains items → list each as "Do NOT use {item}."]
    

    4. Generate brief-preview.html

    Create a single HTML file that visually renders the resolved design tokens. The preview must contain these 4 sections in order:

    1. Color palette swatches — A horizontal row of rectangles, each showing one color from the Color section. Label each with its role (Background, Surface, Text, Accent) and hex code.
    2. Typography specimens — Three text blocks showing Display, Body, and Mono fonts at their declared sizes. Use a sample sentence ("The quick brown fox...") for each.
    3. Spacing ruler — A visual ruler or stacked bars showing section spacing and content padding values, labeled with their px values.
    4. Component preview — Render 2–3 live components (a primary button, a card with title/body, a text input) using the resolved tokens. These should be functional HTML/CSS, not screenshots.

    Style the preview itself with the resolved design system tokens (background color, font, spacing). The preview should look like a design system documentation page.

    5. Report unspecified dimensions

    At the end of output, list any dimensions the user did not specify and the defaults that were applied, including the rule that selected each default:

    Dimensions resolved from defaults:
    - display: set to "same_as_body" (rule: mood=professional_minimal → same_as_body)
    - density: set to "balanced" (rule: static fallback, no spacing preference given)
    - exclude: set to "none" (rule: no constraints unless specified)
    

    This transparency prevents silent assumptions from propagating into the final design.

    Alternatives

    Compare before choosing

    Computed 10042,015

    coreyhaines31/marketingskills

    ab-testing

    When the user wants to plan, design, or implement an A/B test or experiment, or build a growth experimentation program. Also use when the user mentions "A/B test," "split test," "experiment," "test this change," "variant copy," "multivariate test," "hypothesis," "should I test this," "which version is better," "test two versions," "statistical significance," "how long should I run this test," "growth experiments," "experiment velocity," "experiment backlog," "ICE score," "experimentation program

    Computed 10042,015

    coreyhaines31/marketingskills

    churn-prevention

    When the user wants to reduce churn, build cancellation flows, set up save offers, recover failed payments, or implement retention strategies. Also use when the user mentions 'churn,' 'cancel flow,' 'offboarding,' 'save offer,' 'dunning,' 'failed payment recovery,' 'win-back,' 'retention,' 'exit survey,' 'pause subscription,' 'involuntary churn,' 'people keep canceling,' 'churn rate is too high,' 'how do I keep users,' or 'customers are leaving.' Use this whenever someone is losing subscribers o

    Computed 997

    event4u-app/agent-config

    design-review

    Use when the user says "review the design", "check the UI", or wants a comprehensive UI/UX review. Uses a 7-phase methodology covering interaction, responsiveness, accessibility, and more.

    Computed 9631,966

    K-Dense-AI/scientific-agent-skills

    neuropixels-analysis

    Analyze Neuropixels extracellular recordings end-to-end with SpikeInterface. Covers loading SpikeGLX/Open Ephys/NWB data, preprocessing, drift/motion correction, Kilosort4 (and CPU) spike sorting, quality metrics, and unit curation (threshold-based, model-based UnitRefine, and AI-assisted visual review). Use when working with Neuropixels 1.0/2.0 recordings, spike sorting, or extracellular electrophysiology analysis.