flint-theme-author

작성자: microsoft

사용 시점: 브랜드 가이드라인, 웹사이트, 슬라이드 덱, 퍼블리케이션 등에서 커스텀 Flint ThemeSpec을 생성, 번역, 개선, 검토 또는 검증할 때.

npx skills add https://github.com/microsoft/flint-chart --skill flint-theme-author

Flint ThemeSpec authoring

Create a reusable visual system for Flint charts. Translate the user's reference materials into one valid ThemeSpec that can be pasted directly into Theme Lab.

Authoritative references

Read these before authoring. When prose and source disagree, the TypeScript contract is authoritative.

Do not invent keys. If a requested decision cannot be represented by ThemeSpec, omit it and briefly identify the unsupported decision outside the JSON only when the user asked for an explanation.

Output contract

Unless the user asks for commentary, return exactly one valid JSON object:

  • Return the bare ThemeSpec, not { "theme_spec": ... } and not a complete ChartAssemblyInput.
  • Do not wrap JSON in a Markdown fence.
  • Do not include comments, placeholders, ellipses, or trailing commas.
  • Include a stable kebab-case id and a human-readable label.
  • Include only supported fields and only decisions justified by the references.
  • Omit unspecified optional fields rather than guessing.

The top-level output schema is:

{
  "extends": "optional-preset-id",
  "id": "theme-id",
  "label": "Theme label",
  "ink": {},
  "type": {},
  "structure": {},
  "marks": {},
  "labels": {},
  "legend": {},
  "dataLabels": {},
  "annotation": {},
  "furniture": [],
  "facets": {},
  "layout": {},
  "chartDefaults": {},
  "compileDefaults": {},
  "interaction": {},
  "variants": []
}

Every block except id and label may be omitted. The skeleton shows the shape, not a requirement to emit empty blocks.

Theme boundary

A theme controls presentation and compiler behavior across many charts. It may decide color, typography, surfaces, grids, axes, mark geometry, labels, legends, annotations, density, spacing, facets, and reusable chart defaults.

A theme does not choose data fields, encodings, aggregation, filtering, sorting, chart titles, or data values. Never put those decisions in a ThemeSpec. Outside chartDefaults, authored policy must not name a chart type, positional channel, mark type, field, or backend property.

Workflow

  1. Inspect the supplied references. Extract recurring decisions, not isolated decoration from one screenshot.
  2. Ask only for missing decisions that materially affect the theme: identity, required colors, typography, surface/background, density, accessibility constraints, and whether an existing Flint preset is a useful base.
  3. Separate evidence into system roles:
    • surfaces and text hierarchy;
    • categorical, sequential, diverging, and status color;
    • typography roles;
    • axes, grids, baselines, frames, and structural ink;
    • mark weight, corners, points, separators, and spacing;
    • labels, legends, annotations, facets, and layout behavior.
  4. Choose standalone authoring or inheritance.
  5. Author the smallest coherent spec that expresses the system.
  6. Check every key and enum against types.ts.
  7. Return the bare JSON object.

Choosing inheritance

Use extends when a built-in preset already supplies the intended compiler behavior and the new theme is a focused variation. Available preset IDs are:

nyt, economist, swiss, nature, mckinsey, datawrapper, powerbi, powerbi-light, pop, and cartoon.

Nested objects merge. Arrays and scalar values replace the preset value. If you replace ink.series.categorical, also consider replacing categoricalExtended; otherwise high-cardinality charts may return to the base preset's extended palette.

Use a standalone theme when the reference system does not honestly inherit a preset's layout, typography, and presentation behavior.

Schema map

Use the source contract for exhaustive nested fields and enum values. These are the authored blocks and their jobs:

BlockPurpose
inkCanvas, plot, panel, text, structural, series, status, and accent colors
typeMinimum size and role-based typography for headlines, axes, values, annotations, footnotes, and displays
structureSemantic categorical/measure axes, grids, baseline, and frame
marksBand width, strokes, interpolation, opacity, corners, outlines, points, connectors, intervals, summaries, and separators
labelsTruncation, flush behavior, and angle policy
legendVisibility, ordered placement choices, direction, title, swatches, and redundancy suppression
dataLabelsVisibility, placement, and ink behavior for values on marks
annotationUnits, axis titles, number formatting, point emphasis, labels, and statistics
furnitureRepeating masthead tabs and header/footer rules
facetsHeaders, frames, axis repetition, spacing, columns, and scale sharing
layoutDensity, target width, title block, and band step
chartDefaultsOptional defaults keyed by registered chart type or *; caller values still win
compileDefaultsPreferred base size, canvas size, and supported assemble options
interactionTooltip format
variantsConditional policy adaptations; variants may not change ink or type

High-value nested shapes

{
  "ink": {
    "surface": { "source": "house", "canvas": "#ffffff", "plot": "#ffffff", "panel": "#f5f5f5" },
    "text": { "primary": "#111111", "secondary": "#444444", "muted": "#777777", "inverse": "#ffffff" },
    "structure": { "axis": "#222222", "grid": "#dddddd", "frame": "#cccccc", "rule": "#222222", "zero": "#222222", "connector": "#888888" },
    "series": {
      "single": "#0067b8",
      "categorical": ["#0067b8", "#d83b01", "#107c10"],
      "categoricalExtended": ["#0067b8", "#d83b01", "#107c10", "#5c2d91"],
      "overflow": "#777777",
      "sequential": { "stops": ["#e8f3fb", "#0067b8"], "space": "lab", "endpointsAgainstSurface": true },
      "diverging": { "stops": ["#0067b8", "#f5f5f5", "#d83b01"], "neutral": "#f5f5f5", "space": "lab", "endpointsAgainstSurface": true },
      "status": { "positive": "#107c10", "negative": "#d83b01", "neutral": "#777777" }
    },
    "accent": "#0067b8"
  }
}

This is a shape example, not a palette recommendation. Derive actual values from the user's references.

Typography roles accept family, size, weight, style, case, and color. Allowed weights are regular, medium, semibold, and bold. Use available font families or defensible fallback stacks; do not claim a font is available merely because it appears in an image.

Presence-driven policies use omit, hairline, quiet, full, or emphasised. Density uses compact, normal, or airy. Read types.ts before using less common enums.

Translation guidance

  • Map brand neutrals to surfaces, text hierarchy, and structure before mapping accent colors to data series.
  • Treat categorical colors as an ordered identity set. Keep adjacent colors distinguishable and ensure the palette remains legible on the plot surface.
  • Build sequential ramps with ordered lightness. Build diverging ramps around a meaningful neutral. Do not use categorical arrays as numeric ramps.
  • Use status colors only for semantic positive, negative, and neutral meaning.
  • Preserve readable contrast for primary text, axes, labels, and values.
  • Translate recurring geometry into marks; do not encode a one-off shape from a single chart as global policy.
  • Prefer semantic structure (measure versus category) over physical x/y assumptions.
  • Keep the system coherent across line, bar, area, pie, matrix, distribution, multiseries, and faceted charts.
  • Use variants only for real conditional policy and include because to state why the adaptation exists.

Validation checklist

Before returning the JSON, verify:

  • It parses as strict JSON.
  • It is a bare object with id and label.
  • Every top-level and nested key exists in types.ts.
  • Every enum value and value type matches the source contract.
  • It contains no chart data, fields, encodings, titles, or backend JSON.
  • Text and structural ink remain readable against their surfaces.
  • Categorical colors are distinguishable; ramps have ordered lightness and usable endpoints.
  • Inherited arrays are intentionally replaced.
  • The theme expresses reusable rules rather than one chart's decoration.
  • The result can be pasted directly into Flint Theme Lab.

microsoft의 다른 스킬

oss-growth
microsoft
OSS 성장 해커 페르소나
agent-framework-azure-ai-py
microsoft
Microsoft Agent Framework Python SDK(agent-framework-azure-ai)를 사용하여 Azure AI Foundry 에이전트를 구축합니다. AzureAIAgentsProvider로 지속적 에이전트를 만들 때, 호스팅 도구(코드 인터프리터, 파일 검색, 웹 검색)를 사용할 때, MCP 서버를 통합할 때, 대화 스레드를 관리할 때, 또는 스트리밍 응답을 구현할 때 사용합니다. 함수 도구, 구조화된 출력, 다중 도구 에이전트를 다룹니다.
development
airunway-aks-setup
microsoft
Set up AI Runway on AKS — from bare cluster to running model. Covers cluster verification, controller install, GPU assessment, provider setup, and first deployment. WHEN: "setup AI Runway", "onboard AKS cluster", "install AI Runway", "airunway setup", "deploy model to AKS", "GPU inference on AKS", "KAITO setup on AKS", "run LLM on AKS", "vLLM on AKS", "set up model serving on AKS", "AI Runway controller".
devops
appinsights-instrumentation
microsoft
Azure Application Insights로 웹앱을 계측하기 위한 지침입니다. 원격 분석 패턴, SDK 설정, 구성 참조를 제공합니다. WHEN: 앱 계측 방법, App Insights SDK, 원격 분석 패턴, App Insights란 무엇인가, Application Insights 지침, 계측 예시, APM 모범 사례.
devops
applicationinsights-web-ts
microsoft
브라우저/웹 앱을 Application Insights JavaScript SDK(@microsoft/applicationinsights-web)로 계측합니다. Real User Monitoring(RUM) — 페이지 뷰, 클릭, AJAX/fetch 종속성, 예외, 사용자 지정 이벤트, 백엔드 OpenTelemetry 트레이스와 상관관계가 있는 브라우저 측 GenAI 에이전트 트레이스에 사용합니다. SDK Loader Script 및 npm 설정, 프레임워크 확장(React, React Native, Angular), Click Analytics, 텔레메트리 이니셜라이저, 브라우저에서 생성된 에이전트/도구/모델 스팬에 대한 OTel GenAI 의미론적 규칙을 다룹니다.
devops
azure-ai-anomalydetector-java
microsoft
Azure AI Anomaly Detector SDK for Java로 이상 탐지 애플리케이션을 구축하세요. 단변량/다변량 이상 탐지, 시계열 분석 또는 AI 기반 모니터링을 구현할 때 사용하세요.
development
azure-ai-language-conversations-py
microsoft
azure-ai-language-conversations Python SDK를 사용하여 대화형 언어 이해(CLU)를 구현합니다. ConversationAnalysisClient로 대화 의도와 엔터티를 분석하거나, NLP 기능을 구축하거나, 애플리케이션에 언어 이해를 통합할 때 사용합니다.
development
azure-ai-ml-py
microsoft
Azure Machine Learning SDK v2 for Python. ML 작업 영역, 작업, 모델, 데이터 세트, 컴퓨팅 및 파이프라인에 사용합니다. 트리거: "azure-ai-ml", "MLClient", "workspace", "model registry", "training jobs", "datasets".
development